NSIS Working Group                                         Attila Bader
INTERNET-DRAFT                                            Lars Westberg
Intended status: Informational                                 Ericsson
Expires: 13 January 2010                           Georgios Karagiannis
                                                   University of Twente
                                                       Cornelia Kappler
                                                             DeZem GmbH
                                                             Tom Phelan
                                                                  Sonus

                                                           13 July 2009

       RMD-QOSM - The Resource Management in Diffserv QOS Model
                   <draft-ietf-nsis-rmd-15.txt>


Status of this Memo


   This Internet-Draft is submitted to IETF in full conformance with the
   provisions of BCP 78 and BCP 79.

   Internet-Drafts are working documents of the Internet Engineering
   Task Force (IETF), its areas, and its working groups.  Note that
   other groups may also distribute working documents as Internet-
   Drafts.

   Internet-Drafts are draft documents valid for a maximum of six months
   and may be updated, replaced, or obsoleted by other documents at any
   time.  It is inappropriate to use Internet-Drafts as reference
   material or to cite them other than as "work in progress."

   The list of current Internet-Drafts can be accessed at
   http://www.ietf.org/ietf/1id-abstracts.txt.

   The list of Internet-Draft Shadow Directories can be accessed at
   http://www.ietf.org/shadow.html.

   This Internet-Draft will expire on January 13, 2010.


Copyright Notice

   Copyright (c) 2009 IETF Trust and the persons identified as the
   document authors.  All rights reserved.

   This document is subject to BCP 78 and the IETF Trust's Legal
   Provisions Relating to IETF Documents in effect on the date of
   publication of this document (http://trustee.ietf.org/license-info).
   Please review these documents carefully, as they describe your rights
   and restrictions with respect to this document.

Intended status: Informational RFC

 Bader, et al.                                                 [Page 1]


INTERNET-DRAFT                                                 RMD-QOSM


Abstract

   This document describes an NSIS QoS Model for networks that use the
   Resource Management in Diffserv (RMD) concept.  RMD is a technique
   for adding admission control and pre-emption function to
   Differentiated Services (Diffserv) networks.  The RMD QoS Model
   allows devices external to the RMD network to signal reservation
   requests to edge nodes in the RMD network. The RMD Ingress edge nodes
   classify the incoming flows into traffic classes and signals resource
   requests for the corresponding traffic class along the data path to
   the Egress edge nodes for each flow.  Egress nodes reconstitute the
   original requests and continue forwarding them along the data path
   towards the final destination. In addition, RMD defines notification
   functions to indicate overload situations within the domain to the
   edge nodes.


Table of Contents

   1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3
   2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . .5
   3. Overview of RMD and RMD-QOSM . . . . . . . . . . . . . .. . .5
      3.1 RMD . . . . . . . . . . . . . . . . . . . . . . . . . . .6
      3.2 Basic features of RMD-QOSM . . . . . . . . . . . . . . . 9
          3.2.1 Role of the QNEs . . . . . . . .. . . . . . . . . .9
          3.2.2 RMD-QOSM/QoS-NSLP signaling . . . . . . . . . . . 10
          3.2.3 RMD-QOSM Applicability and considerations. . . . .12
   4. RMD-QOSM, Detailed Description . . . . . . . . . . . .. . . 13
      4.1 RMD-QSpec Definition . . . . . . . . . . . . . . . . . .13
          4.1.1 RMD-QOSM object contribution . . . . . . . . . . .14
          4.1.2 PHR Container . . . . . . . . . . . . . . . . . . 15
          4.1.3 PDR Container  . . . . . . . . . . . . . . . . . .16
      4.2 Message format . . . . . . . . . . . . . . . . . . . . .19
      4.3 RMD node state management . . . . . . . . . . . . . . . 19
          4.3.1 Aggregated versus per flow reservations at the
                QNE edges . . . . . . . . . . . . . . . . . . . . 19
          4.3.2 Measurement-based method . . . . . . . . . . . . .21
          4.3.3 Reservation-based method . .. . . . . . . . . . . 23
      4.4 Transport of RMD-QOSM messages . . . . . . . . . . . . .24
      4.5 Edge discovery and addressing of messages . . . . . . . 26
      4.6 Operation and sequence of events . . . . . . . . . . . .27
          4.6.1 Basic unidirectional operation . . . . . . . . . .27
             4.6.1.1 Successful reservation. . . . . . . . . . . .28
             4.6.1.2 Unsuccessful reservation . . . . . . . . . . 38

 Bader, et al.                                                  [Page 2]


INTERNET-DRAFT                                                 RMD-QOSM


             4.6.1.3 RMD refresh reservation. . . . . . . . . . . 42
             4.6.1.4 RMD modification of aggregated reservation . 45
             4.6.1.5 RMD release procedure. . . . . . . . . . . . 46
             4.6.1.6 Severe congestion handling  . . . . . . . . .53
             4.6.1.7 Admission control using congestion
                     notification based on probing . . . . . .  . 59
          4.6.2 Bidirectional operation . . . . . . . . . . . . . 62
             4.6.2.1 Successful and unsuccessful reservation . . .64
             4.6.2.2 Refresh reservation . . . . . . . . . . . . .67
             4.6.2.3 Modification of aggregated reservation . . . 68
             4.6.2.4 Release procedure . . . . . . . . . . . . . .69
             4.6.2.5 Severe congestion handling . . . . . . . . . 69
             4.6.2.6 Admission control using congestion
                     notification based on probing . . . . . . . .72
      4.7 Handling of additional errors . . . . . . . . . . . . . 74
   5. Security Consideration. . . . . . . . . . . . . . . . . . . 75
   6. IANA Considerations. . . . . . . . . . . . . . . . . . . . .78
   7. Acknowledgments. . . . . . . . . . . . . . . . . . . . . . .78
   8. Authors` Addresses. . . . . . . . . . . . . . . . . . . . . 79
   9. Normative References . . . . . . . . . . . . . . . . . . . .80
   10. Informative References . . . . . . . . . . . . . . . . . . 80


1.  Introduction

   This document describes a Next Steps In Signaling (NSIS) QoS model
   for networks that use the Resource Management in Diffserv (RMD)
   framework ([RMD1], [RMD2], [RMD3], [RMD4]). RMD adds admission
   control to Diffserv networks and allows nodes external to the
   networks to dynamically reserve resources within the Diffserv
   domains.

   The Quality of Service NSIS Signaling Layer Protocol (QoS-NSLP)
   [QoS-NSLP] specifies a generic protocol for carrying Quality of
   Service(QoS) signaling information end-to-end in an IP network.
   Each network along the end-to-end path is expected to implement a
   specific QoS Model (QOSM) specified by the QSpec template [QSP-T]
   that interprets the requests and installs the necessary mechanisms,
   in a manner that is appropriate to the technology in use in the
   network, to ensure the delivery of the requested QoS.

Bader, et al.                                                  [Page 3]


INTERNET-DRAFT                                                 RMD-QOSM

   This document specifies an NSIS QoS Model for RMD networks (RMD-
   QOSM), and an RMD-specific QSpec (RMD-QSPec) for expressing
   reservations in a suitable form for simple processing by internal
   nodes.  They are used in combination with the QoS-NSLP to provide
   QoS signaling service in an RMD network.  Figure 1 shows an RMD
   network with the respective entities.

                          Stateless or reduced state        Egress
   Ingress                RMD nodes                         Node
   Node                   (Interior Nodes; I-Nodes)        (Stateful
   (Stateful              |          |            |         RMD QoS
   RMD QoS NLSP           |          |            |         NSLP Node)
   Node)                  V          V            V
   +-------+   Data +------+      +------+       +------+     +------+
   |-------|--------|------|------|------|-------|------|---->|------|
   |       |   Flow |      |      |      |       |      |     |      |
   |Ingress|        |I-Node|      |I-Node|       |I-Node|     |Egress|
   |       |        |      |      |      |       |      |     |      |
   +-------+        +------+      +------+       +------+     +------+
            =================================================>
            <=================================================
                                  Signaling Flow

   Figure 1: Actors in the RMD-QOSM

   Internally to the RMD network, RMD-QOSM together with QoS-NSLP
   [QoS-NSLP] defines a scalable QoS signaling model in which per-flow
   QoS-NSLP and NTLP states are not stored in Interior nodes but
   per-flow signaling is performed (see [QoS-NSLP]).

   In the RMD-QOSM, only routers at the edges of a Diffserv domain
   (Ingress and Egress nodes) support the (QoS-NSLP) stateful
   operation, see Section 4.7 of [QoS-NSLP]. Interior nodes support
   either the(QoS-NSLP) stateless operation, or a reduced-state
   operation with coarser granularity than the edge nodes.

   After the terminology in Section 2, we give an overview of RMD and
   the RMD-QOSM in Section 3.  In Section 4 we give a detailed
   description of the RMD-QOSM, including the role of QNEs, the
   definition of the QSpec, mapping of QSpec generic parameters onto
   RMD-QOSM parameters, state management in QNEs, and operation and
   sequence of events.  Section 5 discusses security issues.

Bader, et al.                                                  [Page 4]


INTERNET-DRAFT                                                 RMD-QOSM

2.  Terminology

   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
   "SHOULD", "SHOULD NOT", "RECOMMENDED",  "MAY", and "OPTIONAL" in
   this document are to be interpreted as described in [RFC2119].

   The terminology defined by GIST [GIST] and QoS-NSLP [QoS-NSLP]
   applies to this draft.

   In addition, the following terms are used:

   NSIS domain: a NSIS signalling capable domain.

   RMD domain: a NSIS domain that is capable of supporting the RMD-QOSM
   signalling and operations.

   Edge node: a QoS-NSLP node on the boundary of some
   administrative domain that connects one NSIS domain to a node
   either in another NSIS domain or in a non NSIS domain.

   NSIS aware node: a node that is aware of NSIS signalling and RMD-
   QOSM operations, such as severe congestion detection and DSCP
   marking.

   NSIS unaware: a node that is unware of NSIS signalling, but is
   aware of RMD-QOSM operations such as severe congestion detection
   and DSCP marking.

   Ingress node: An edge node in its role in handling the traffic
   as it enters the NSIS domain.

   Egress node: An edge node in its role in handling the traffic
   as it leaves the NSIS domain.
   Interior node: a node in a NSIS domain that is not an edge node.

   Congestion: is a temporal network state that occurs when the traffic
   (or when traffic associated with a particular PHB) passing through a
   link is slightly higher than the capacity allocated for the link (or
   allocated for the particular PHB). If no measures are taken than the
   traffic passing through this link may temporarily slightly degrade in
   QoS. This type of congestion is usually solved using admission
   control mechanisms.

   Severe congestion: is a congestion that occurs when a node or link
   fails and the traffic is rerouter through another node or link. If no
   measures are taken than the node or the link can become severely
   congested and all traffic passing through the node or link will
   severely degrade in QoS. This type of congestion cannot be solved
   using admission control mechanisms.


Bader, et al.                                                  [Page 5]


INTERNET-DRAFT                                                 RMD-QOSM

3.  Overview of RMD and RMD-QOSM

3.1.  RMD

   The Differentiated Services (Diffserv) architecture ([RFC2475],
   [RFC2638]) was introduced as a result of efforts to avoid the
   scalability and complexity problems of Intserv [RFC1633].
   Scalability is achieved by offering services on an aggregate
   rather than per-flow basis and by forcing as much of the per-flow
   state as possible to the edges of the network.  The service
   differentiation is achieved using the Differentiated Services (DS)
   field in the IP header and the Per-Hop Behavior (PHB) as the main
   building blocks.  Packets are handled at each node according to the
   PHB indicated by the DS field in the message header.

   The Diffserv architecture does not specify any means for devices
   outside the domain to dynamically reserve resources or receive
   indications of network resource availability.  In practice, service
   providers rely on short active time Service Level Agreements (SLAs)
   that statically define the parameters of the traffic that will be
   accepted from a customer.

   RMD was introduced as a method for dynamic reservation of resources
   within a Diffserv domain.  It describes a method that is able to
   provide admission control for flows entering the domain and a
   congestion handling algorithm that is able to terminate flows in
   case of congestion due to a sudden failure (e.g., link, router)
   within the domain.

   In RMD, scalability is achieved by separating a fine-grained
   reservation mechanism used in the edge nodes of a Diffserv domain
   from a much simpler reservation mechanism needed in the Interior
   nodes.  Typically it is assumed that edge nodes support per-
   flow QoS states in order to provide QoS guarantees for each flow.
   Interior nodes use only one aggregated reservation state per traffic
   class or no states at all.  In this way it is possible to handle
   large numbers of flows in the Interior nodes. Furthermore, due to
   the limited functionality supported by the Interior nodes, this
   solution allows fast processing of signaling messages.

   The possible RMD-QOSM applicabilities are described in Section
   3.2.3. Two main basic admission control modes are supported:
   reservation-based and measurement-based admission control that can
   be used in combination with a severe congestion handling solution.
   The severe congestion handling solution is used in the situation
   that a link/node becomes severely congested due to the fact that the
   traffic supported by a failed link/node is rerouted and has to be
   processed by this link/node. Furthermore, RMD-QOSM supports both
   uni-directional and bi-directional reservations.

Bader, et al.                                                  [Page 6]


INTERNET-DRAFT                                                 RMD-QOSM

   Another important feature of RMD-QOSM is that the intra-domain
   sessions supported by the edges can be either per flow sessions or
   per aggregate sessions. In case of the per flow intra-domain
   sessions, the maintained per flow intra-domain states have a one-to-
   one dependency to the per flow end-to-end states supported by the
   same edge. In case of the per-aggregate sessions the maintained per-
   aggregate states have a one-to-many relationship to the per flow
   end-to-end states supported by the same edge.

   In the reservation-based method, each Interior node maintains
   only one reservation state per traffic class.  The Ingress edge
   nodes aggregate individual flow requests into PHB traffic classes,
   and signal changes in the class reservations as necessary.  The
   reservation is quantified in terms of resource units (or bandwidth).
   These resources are requested dynamically per PHB and reserved on
   demand in all nodes in the communication path from an Ingress node
   to an Egress node.

   The measurement-based algorithm continuously measures traffic levels
   and the actual available resources, and admits flows whose resource
   needs are within what is available at the time of the request. Once
   an admission decision is made, no record of the decision need be
   kept at the interior nodes.  The advantage of measurement-based
   resource management
   protocols is that they do not require pre-reservation state nor
   explicit release of the reservations at the interior nodes.
   Moreover, when the user
   traffic is variable, measurement based admission control could
   provide higher network utilization than, e.g., peak-rate
   reservation.  However, this can introduce an uncertainty in the
   availability of the resources.

   Two types of measurement based admission control schemes are
   possible:

   * Congestion notification function based on probing:

   This method can be used to implement a simple measurement-based
   admission control within a Diffserv domain. In this scenario the
   interior nodes are not NSIS aware nodes. In these interior nodes
   thresholds are set for the traffic belonging to different PHBs in
   the measurement based admission control function. In this scenario
   an end-to-end NSIS message is used as a probe packet, meaning that
   the DSCP field in the header of the IP packet that carries the NSIS
   message is re-marked when the predefined congestion threshold is
   exceeded. Note that when the predefined congestion threshold is
   exceeded all packets are remarked by a node, including NSIS
   messages. In this way the edges can admit or reject flows that are
   requesting resources. The rate of the re-marked data packets is used
   to detect a congestion situation that can influence the admission
   control decisions.

Bader, et al.                                                  [Page 7]


INTERNET-DRAFT                                                 RMD-QOSM

   * NSIS measurement-based admission control:

   In this case the measurement-based admission control functionality
   is implemented in NSIS aware stateless routers. The main difference
   between this type of admission control and the congestion
   notification based on probing is related to the fact that this type
   of admission control is applied mainly on NSIS aware nodes, giving
   the possibility to apply measuring techniques, see e.g., [JaSh97],
   [GrTs03], that are using current and past information on NSIS
   sessions that requested resources from an NSIS aware interior node.
   The admission decision is positive if the currently carried traffic,
   as characterized by the measured statistics, plus the requested
   resources for the new flow exceeds the system capacity with a
   probability smaller than some alpha. Otherwise, the admission
   decision is negative.


   RMD describes the following procedures:

   * Classification of an individual resource reservation or a resource
     query into Per Hop Behavior (PHB) groups at the Ingress node of
     the domain,

   * Hop-by-hop admission control based on a PHB within the
     domain. There are two possible modes of operation for internal
     nodes to admit requests. One mode is the stateless or
     measurement-based mode, where the resources within the domain are
     queried. Another mode of operation is the reduced-state
     reservation or reservation based mode, where the resources within
     the domain are reserved.

   * a method to forward the original requests across the domain up to
     the Egress node and beyond.

   * a congestion control algorithm that notifies the egress edge nodes
     about congestion. It is able to terminate the appropriate number
     of flows in case a of congestion due to a sudden failure (e.g.,
     link or router failure) within the domain.


3.2. Basic features of RMD-QOSM

3.2.1 Role of the QNEs

   The protocol model of the RMD-QOSM is shown in Figure 2.  The figure
   shows QNI and QNR nodes, not part of the RMD network, that are the
   ultimate initiator and receiver of the QoS reservation requests.  It
   also shows QNE nodes that are the Ingress and Egress nodes in the
   RMD domain (QNE Ingress and QNE Egress), and QNE nodes that are
   Interior nodes (QNE Interior).

Bader, et al.                                                  [Page 8]


INTERNET-DRAFT                                                 RMD-QOSM

   All nodes of the RMD domain are usually QoS-NSLP aware nodes.
   However, in the scenarios where the congestion notification function
   based on probing is used, then the interior nodes are not NSIS
   aware.  Edge nodes store and maintain QoS-NSLP and NTLP states and
   therefore are stateful nodes.  The NSIS aware Interior nodes are
   NTLP stateless. Furthermore they are either QoS-NSLP stateless (for
   NSIS measurement-based operation), or are reduced state nodes
   storing per PHB aggregated QoS-NSLP states (for reservation-based
   operation).

   Note that the RMD domain MAY contain Interior nodes that are
   not NSIS aware nodes (not shown in the figure).  These nodes are
   assumed to have sufficient capacity for flows that might be
   admitted.  Furthermore, some of these NSIS unaware nodes MAY be used
   for measuring the traffic congestion level on the data path. These
   measurements can be used by RMD-QOSM in the congestion control based
   on probing operation and/or severe congestion operation
   (see Section 4.6.1.6).

     |------|   |-------|                           |------|   |------|
     | e2e  |<->| e2e   |<------------------------->| e2e  |<->| e2e  |
     | QoS  |   | QoS   |                           | QoS  |   | QoS  |
     |      |   |-------|                           |------|   |------|
     |      |   |-------|   |-------|   |-------|   |------|   |      |
     |      |   | local |<->| local |<->| local |<->| local|   |      |
     |      |   | QoS   |   |  QoS  |   |  QoS  |   |  QoS |   |      |
     |      |   |       |   |       |   |       |   |      |   |      |
     | NSLP |   | NSLP  |   | NSLP  |   | NSLP  |   | NSLP |   | NSLP |
     |st.ful|   |st.ful |   |st.less/   |st.less/   |st.ful|   |st.ful|
     |      |   |       |   |red.st.|   |red.st.|   |      |   |      |
     |      |   |-------|   |-------|   |-------|   |------|   |      |
     |------|   |-------|   |-------|   |-------|   |------|   |------|
     ------------------------------------------------------------------
     |------|   |-------|   |-------|   |-------|   |------|   |------|
     | NTLP |<->| NTLP  |<->| NTLP  |<->| NTLP  |<->| NTLP |<->|NTLP  |
     |st.ful|   |st.ful |   |st.less|   |st.less|   |st.ful|   |st.ful|
     |------|   |-------|   |-------|   |-------|   |------|   |------|
       QNI         QNE        QNE         QNE          QNE       QNR
     (End)     (Ingress)   (Interior)  (Interior)   (Egress)    (End)

         st.ful: stateful, st.less: stateless
         st.less red.st.: stateless or reduced state

   Figure 2: Protocol model of stateless/reduced state operation

Bader, et al.                                                  [Page 9]


INTERNET-DRAFT                                                 RMD-QOSM


3.2.2 RMD-QOSM/QoS-NSLP Signaling

   The basic RMD-QOSM/QoS-NSLP signaling is shown in Figure 3. The
   signalling scenarios are accomplished using the QoS-NSLP processing
   rules defined in [QoS-NSLP], in combination with the RMF triggers
   sent via the QoS-NSLP-RMF API described in [QoS-NSLP]. A RESERVE
   message is created by a QNI with an Initiator QSpec describing the
   reservation and forwarded along the path towards the QNR.  When the
   original RESERVE message arrives at the Ingress node, an RMD-QSpec
   is constructed based on the initial QSpec in the message (usually
   the Initiator QSpec).  The RMD-QSpec is sent in a intra-domain,
   independent RESERVE message through the Interior nodes towards the
   QNR. This intra-domain RESERVE message uses the  GIST datagram
   signaling mechanism. Note that the RMD-QOSM cannot directly specify
   that the GIST datagram mode SHOULD be used. This can however be
   notified by using the GIST API Transfer-Attributes, such as
   unreliable, low level of security and use of local policy.
   Meanwhile, the original RESERVE message is sent to the Egress node
   on the path to the QNR using the reliable transport mode of NTLP.


          QNE Ingress     QNE Interior     QNE Interior   QNE Egress
        NTLP stateful  NTLP stateless  NTLP stateless  NTLP stateful
               |               |               |              |
       RESERVE |               |               |              |
      -------->| RESERVE       |               |              |
               +--------------------------------------------->|
               | RESERVE`      |               |              |
               +-------------->|               |              |
               |               | RESERVE`      |              |
               |               +-------------->|              |
               |               |               | RESERVE`     |
               |               |               +------------->|
               |               |               |     RESPONSE`|
               |<---------------------------------------------+
               |               |               |              | RESERVE
               |               |               |              +------->
               |               |               |              |RESPONSE
               |               |               |              |<-------
               |               |               |     RESPONSE |
               |<---------------------------------------------+
       RESPONSE|               |               |              |
      <--------|               |               |              |

   Figure 3: Sender-initiated reservation with Reduced State Interior
             Nodes

Bader, et al.                                                 [Page 10]


INTERNET-DRAFT                                                 RMD-QOSM

   Each QoS-NSLP node on the data path processes the intra-domain
   RESERVE message and checks the availability of resources with either
   the reservation-based or the measurement-based method.  When the
   message reaches the Egress node, and the reservation is successful
   in each Interior node, an intra-domain (local) RESPONSE` is sent
   towards the ingress node and the original (end-to-end) RESERVE
   message is forwarded to the next domain.  When the Egress node
   receives a RESPONSE message from the downstream end, it is forwarded
   directly to the Ingress node.

   If an intermediate node cannot accommodate the new request, it
   indicates this by marking a single bit in the message, and continues
   forwarding the message until the Egress node is reached. From the
   Egress node an intra-domain RESPONSE` and an original RESPONSE
   message are sent directly to the Ingress node.

   As a consequence in the stateless/reduced state domain only sender-
   initiated reservation can be performed and functions requiring per
   flow NTLP or QoS-NSLP states, like summary and reduced refreshes,
   cannot be used. If per flow identification, is needed, i.e.,
   associating the flow IDs for the reserved resources, Edge nodes act
   on behalf of Interior nodes.


3.2.3 RMD-QOSM Applicability and considerations

   The RMD-QOSM is a Diffserv-based bandwidth management methodology
   that is not able to provide a full Diffserv support. The reason of
   this is that the RMD-QOSM concept can only support the (Expedited
   Forwarding) EF-like functionality behavior, but is not able to
   support the full set of (Assured Forwarding) AF-like functionality.
   The bandwidth information REQUIRED by the EF-like functionality
   behaviour can be supported by RMD-QOSM carrying the bandwidth
   information in the <QoS Desired> parameter, see [QSP-T]. The full set
   of (Assured Forwarding) AF-like functionality requires information
   that is specified in two token buckets. The RMD-QOSM is not
   supporting the use of two token buckets and therefore, it is not able
   to support the full set of AF-functionality. Note however,
   that RMD-QOSM could also support a single AF PHB, when the traffic
   or the upper limit of the traffic can be characterized by a single
   bandwidth parameter.

   A very important consideration on using RMD-QOSM is that within one
   RMD domain only one of the following RMD-QOSM schemes can be used at
   a time. Thus a RMD router can never process and use two different
   RMD-QOSM signaling schemes at the same time. The operator of an RMD
   domain has to pre-configure all routers in the domain such that
   within one RMD domain only one of the below described RMD-QOSM
   schemes can be used at a time. Note however, that there is no single
   to implement scheme variant.

Bader, et al.                                                 [Page 11]


INTERNET-DRAFT                                                 RMD-QOSM

   The congestion situations, see Section 2, are solved using admission
   control mechanism, e.g., "per flow congestion notification based on
   probing", while the severe congestion situations, see Section 2, are
   solved using the severe congestion handling mechanisms, e.g., "Severe
   congestion handling by proportional data packet marking".

   The RMD domain MUST be engineered in such a way that RMD-QOSM
   messages could be transported using the GIST Query and
   Data messages in Q-mode, see [GIST]. This means that the Path MTU
   MUST be engineered in such a way that the RMD-QOSM message are
   transported without fragmentation. Furthermore, the RMD domain MUST
   be engineered in such a way to guarantee capacity for the GIST
   Query and Data messages in Q-mode, within the rate control limits
   imposed by GIST, see [GIST].

   The RMD domain has to be configured such that the GIST context-free
   flag (C-flag) MUST be set (C=1) for Query messages and Data
   messages sent in Q-mode, see [GIST].

   It is important to note that the concepts described in Sections
   4.6.1.6.2, 4.6.2.5.2, 4.6.1.6.2 and 4.6.2.5.2 contributed to
   the PCN WG standardisation.

   The available RMD-QOSM/QoS-NSLP signaling schemes are:

   * "per flow congestion notification based on probing" (see
     Sections 4.3.2, 4.6.1.7., 4.6.2.6.). Note that this scheme uses for
     severe congestion handling the "Severe congestion handling by
     proportional data packet marking", see Section 4.6.1.6.2,
     4.6.2.5.2). Furthermore, the interior nodes are considered to be
     Diffserv aware, but NSIS unaware nodes, see Section 4.3.2.

   * "per flow RMD NSIS measurement based admission control" (see
     Sections 4.3.2, 4.6.1, 4.6.2). Note that this scheme uses for
     severe congestion handling the "Severe congestion handling by
     proportional data packet marking", see Section 4.6.1.6.2,
     4.6.2.5.2). Furthermore, the interior nodes are considered to be
     NSIS aware nodes, see Section 4.3.2.

   * "per flow RMD reservation based" in combination with "severe
     congestion handling by the RMD-QOSM refresh procedure" (see
     Sections 4.3.3, 4.6.1, 4.6.1.6.1, 4.6.2.5.1). Note that this
     scheme uses for severe congestion handling the "Severe congestion
     handling by the RMD-QOSM refresh" procedure, see Section 4.6.1.6.1,
     4.6.2.5.1). Furthermore, the intra-domain sessions supported by the
     edge nodes are per flow sessions, see Section 4.3.3.

Bader, et al.                                                 [Page 12]


INTERNET-DRAFT                                                 RMD-QOSM

   * "per flow RMD reservation based" in combination with "severe
     congestion handling by proportional data packet marking" procedure
     (see Sections 4.3.3, 4.6.1, 4.6.1.6.2, 4.6.2.5.2). Note that
     this scheme uses for severe congestion handling the "Severe
     congestion handling by proportional data packet marking" procedure,
     see Section 4.6.1.6.2, 4.6.2.5.2). Furthermore, the intra-domain
     sessions supported by the edge nodes are per flow sessions, see
     Section 4.3.3.

   * "per aggregate RMD reservation based" in combination with
     "severe congestion handling by the RMD-QOSM refresh procedure" (see
     Sections 4.3.1, 4.6.1, 4.6.1.6.1, 4.6.2.5.1). Note that this
     scheme uses for severe congestion handling the "Severe congestion
     handling by the RMD-QOSM refresh" procedure, see Section 4.6.1.6.1,
     4.6.2.5.1). Furthermore, the intra-domain sessions supported by the
     edge nodes are per aggregate sessions, see Section 4.3.1.

   * "per aggregate RMD reservation based" in combination with
     "severe congestion handling by proportional data packet marking"
     procedure (see Sections 4.3.1, 4.6.1, 4.6.1.6.2, 4.6.2.5.2).
     Note that this scheme uses for severe congestion handling the
     "Severe congestion handling by proportional data packet marking"
     procedure, see Section 4.6.1.6.2, 4.6.2.5.2). Furthermore, the
     intra-domain sessions supported by the edge nodes are per aggregate
     sessions, see Section 4.3.1.

4.  RMD-QOSM, Detailed Description

   This section describes the RMD-QOSM in more detail.  In particular,
   it defines the role of stateless and reduced-state QNEs, the
   RMD-QOSM QSpec Object, the format of the RMD-QOSM QoS-NSLP messages
   and how QSpecs are processed and used in different protocol
   operations.


4.1.  RMD-QSpec Definition

   The RMD-QOSM uses the QSpec format specified in [QSP-T].
   The <I> flag is set to "Local" (i.e., "1") and the <Qspec Proc> is
   set as follows:
   * Message Sequence = 0: Sender initiated
   * Object combination = 1: <QoS Desired> for RESERVE and
     <QoS Reserved> for RESPONSE

   The <QSPEC Version> used by RMD-QOSM is the default version, see
  [QSP-T]. The <QSPEC Type> used by the RMD-QOSM is
   assigned by IANA, see Section 6. The <Traffic Handling Directives>
   contains the following fields:

   <Traffic Handling Directives> = <PHR container> <PDR container>

Bader, et al.                                                 [Page 13]


INTERNET-DRAFT                                                 RMD-QOSM

   The Per Hop Reservation container (PHR container) and
   the Per Domain Reservation container (PDR container) are specified
   in Section 4.1.2 and 4.1.3, respectively. The <PHR container>
   contains the traffic handling directives for intra-domain
   communication and reservation.  The <PDR container> contains

   additional traffic handling directives that is needed for
   edge-to-edge communication. The parameter IDs used by the <PHR
   container> and <PDR container> are assigned by IANA, see Section 6.
   For clarity Reasons we will assigned temporarily, the following names
   to the PHR and PDR containers:
       * PHR_1 to PHR_3 for the <PHR container>
       * PDR_4 to PDR_10 for the <PDR container>
   After IANA assigns the proper ID values to the PHR and PDR
   containers, then the above list has to be replaced accordingly.

   The "RMD-QOSM object combination", i.e., <QoS Desired> and
   <QoS Reserved>,  is specified in Section 4.1.1. The "RMD-
   QOSM QoS object combination" and the <PHR container> are used and
   processed by the Edge and Interior nodes.  The <PDR container> field
   is only processed by Edge nodes.


4.1.1.  RMD-QOSM object combination

   The "RMD-QOSM object combination" carried by the RESERVE message only
   contains the QoS Desired object [QSP-T]. The QoS Reserved object is
   carried by the RESPONSE message.

   "RMD-QOSM object combination" = <QoS Desired> for RESERVE
   "RMD-QOSM object combination" = <QoS Reserved> for RESPONSE
   <QoS Desired> = <Bandwidth> <PHB Class> <Admission Priority>
   <QoS Reserved> = <Bandwidth> <PHB Class> <Admission Priority>

   The bit format of the <PHB Class> (see [QSP-T] and Figure 4 and
   Figure 5) and <Admission Priority> complies to the bit format
   specified in [QSP-T]. The bit format used for the <Bandwidth>
   parameter is shown below and it is identical to the peak rate (p)
   <TMOD-1> parameter format specified in [QSP-T]. Note that after IANA
   assigns the proper ID value to the <Bandwdith> parameter then the
   Bandwdith_ID term has to be replaced accordingly.

     0                   1                   2                   3
     0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
    |1|E|0|r|       Bandwidth_IDID    |r|r|r|r|          1            |
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
    | Peak Data Rate-1 (p)  (32-bit IEEE floating point number)     |
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

            <Bandwidth> parameter format

Bader, et al.                                                 [Page 14]


INTERNET-DRAFT                                                 RMD-QOSM

   Note that for the RMD-QOSM a reservation established without an
   <Admission Priority> parameter is equivalent to a reservation
   established with an <Admission Priority> whose value is 1.

            0                   1
            0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5
           +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
           | DSCP      |0 0 0 0 0 0 0 0 X 0|
           +---+---+---+---+---+---+---+---+

             Figure 4: DSCP parameter

            0                   1
            0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5
           +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
           |    PHB ID code        |0 0 X X|
           +---+---+---+---+---+---+---+---+

            Figure 5: PHB ID Code parameter


4.1.2.  PHR Container

   This section describes the parameters used by the PHR container,
   which are used by the RMD-QOSM functionality available at the
   Interior nodes.

   <PHR container> = <Overload %>, <K>, <S>,<M>,
   <Admitted Hops>, <B>, <Hop_U>, <Time Lag>, <Max Admitted Hops>

   The bit format of the PHR container can be seen in Figure 6. Note
   that in Figure 6 <Hop U> is represented as <U>. Furthermore, in
   Figure 6 <Max Admitted Hops> is represented as <Max Adm Hops>.

     0                   1                   2                   3
     0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
    |M|E|N|r|       Container ID    |r|r|r|r|          2            |
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
    |S|M| Admitted  Hops|B|U| Time  Lag     |  Overload %   |K|     |
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
    | Max Adm  Hops |                                               |
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

        Figure 6: PHR container

   Parameter/Container ID: 8 bit field, indicating the PHR type:
   PHR_Resource_Request, PHR_Release_Request, PHR_Refresh_Update.

   "PHR_Resource_Request" (Container ID = PHR_1): initiate or update
   the traffic class reservation state on all nodes located on the
   communication path between the QNE(Ingress) and QNE(Egress) nodes.

Bader, et al.                                                 [Page 15]


INTERNET-DRAFT                                                 RMD-QOSM

   "PHR_Refresh_Update" (Container ID = PHR_2): refresh the
   traffic class reservation soft state on all nodes located on the
   communication path between the QNE(Ingress) and QNE(Egress)
   nodes according to a resource reservation request that was
   successfully processed during a previous refresh period.

   "PHR_Release_Request" (Container ID = PHR_3): explicitly release, by
   subtraction, the reserved resources for a particular flow
   from a traffic class reservation state.

   <S> (Severe Congestion):
   1 bit.  In case of a route change refreshing RESERVE messages
   follow the new data path, and hence resources are requested
   there.  If the resources are not sufficient to accommodate the new
   traffic severe congestion occurs.  Severe congested Interior nodes
   SHOULD notify Edge QNEs about the congestion by setting the S bit.

   <Overload %>:
   8 bits In case of severe congestion the level of overload is
   indicated by the Overload %.  Overload % SHOULD be higher than 0 if
   S bit is set.  If overload in a node is greater than the overload
   in a previous node then Overload % SHOULD be updated. For more
   details see Section 4.6.1.6.1. Note that this field represents a real
   parameter.

   <M>:
   1 bit.  In case of unsuccessful resource reservation or resource
   query in an Interior QNE, this QNE sets the M bit in order to
   notify the Egress QNE.

   <Admitted Hops>:
   8 bit field.  The <Admitted Hops> counts the number of hops in the
   RMD domain where the reservation was successful.  The <Admitted
   Hops> is set to "0" when a RESERVE message enters a domain and it
   MUST be incremented by each Interior QNE, provided that the Hop_U bit
   is not set.  However when a QNE that does
   not have sufficient resources to admit the reservation is reached,
   the M Bit is set, and the <Admitted Hops> value is frozen, by setting
   the Hop_U bit to "1". Note that the <Admitted Hops> parameter in
   combinations with the <Max Admitted Hops> and <K> parameters are used
   during the RMD partial release procedures, see Section 4.6.1.5.2.

   <Hop_U> (NSLP_Hops unset):
   1-bit. The QNE(Ingress) node MUST set the <Hop_U> parameter to
   0.  This parameter SHOULD be set to "1" by a node when the node does
   not increase the <Admitted Hops> value. This is the case when an
   RMD-QOSM reservation-based node is not admitting the reservation
   request. When <Hop_U> is set "1" the <Admitted Hops> SHOULD NOT be
   changed. Note that this flag in combination with the <Admitted Hops>
   flag are used to locate the last node that successfully processed a
   reservation request, see Section 4.6.1.2.

   <B>: 1 bit.  When set to "1" it indicates bi-directional reservation.

Bader, et al.                                                 [Page 16]


INTERNET-DRAFT                                                 RMD-QOSM

   <Time Lag>: 8 bit field.  The time lag used in a sliding window
   over the refresh period.

   <K>: 1 bit. When set to "1" it indicates that the resources/bandwidth
   carried by a tearing RESERVE MUST NOT be released and the
   resources/bandwidth carried by a non tearing RESERVE MUST NOT be
   reserved/refreshed.

   <Max Admitted Hops>:
   8-bit.  The <Admitted Hops> value that has been carried by the
   PHR container field used to identify the RMD reservation based node
   that admitted or process a "PHR_Resource_Request"


4.1.3.  PDR container

   This section describes the parameters of the PDR container, which are
   used by the RMD-QOSM functionality available at the
   Edge nodes.

   The bit format of the PDR container can be seen in Figure 7.

   <PDR container> = <Overload %>  <S> <M> <Max
   Admitted Hops> <B> [<PDR Bandwidth>]

   Note that in Figure 7 <Max Admitted Hops> is represented as
   <Max Adm Hops>.

       0                   1                   2                   3
       0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
      |M|E|N|r|   Container ID        |r|r|r|r|          2            |
      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
      |S|M| Max Adm  Hops |B|  Overload %   |       Empty       |     |
      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
      |PDR Bandwidth(32-bit IEEE floating point.number)               |
      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

        Figure 7: PDR container

   Parameter/Container ID:

   8-bit field identifying the type of PDR container field.

   "PDR_Reservation_Request" (Parameter/Container ID = PDR_4):
   Generated by the QNE(Ingress) node in order to initiate or update
   the QoS-NSLP per domain reservation state in the QNE(Egress) node

   "PDR_Refresh_Request" (Parameter/Container ID = PDR_5): generated by
   the QNE(Ingress) node and sent to the QNE(Egress) node to refresh,
   in case needed, the QoS-NSLP per domain reservation states
   located in the QNE(Egress) node

Bader, et al.                                                 [Page 17]


INTERNET-DRAFT                                                 RMD-QOSM

   "PDR_Release_Request" (Parameter/Container ID = PDR_6): generated
   and sent by the QNE(Ingress) node to the QNE(Egress) node to release
   the per domain reservation states explicitly


   "PDR_Reservation_Report" (Parameter/Container ID = PDR_7): generated
   and sent by the QNE(Egress) node to the QNE(Ingress) node to
   report that a "PHR_Resource_Request" and a
   "PDR_Reservation_Request" traffic handling directive fields have been
   received and that the request has been admitted or rejected

   "PDR_Refresh_Report" (Parameter/Container ID = PDR_8) generated and
   sent by the QNE(Egress) node in case needed, to the QNE(Ingress)
   node to report that a "PHR_Refresh_Update" traffic handling directive
   field has been received and has been processed

   "PDR_Release_Report" (Parameter/Container ID = PDR_9) generated and
   sent by the QNE(Egress) node in case needed, to the QNE(Ingress)
   node to report that a "PHR_Release_Request" and a
   "PDR_Release_Request" traffic handling directive fields have been
   received and have been processed.

   "PDR_Congestion_Report" (Parameter/Container ID = PDR_10): generated
   and sent by the QNE(Egress) node to the QNE(Ingress) node and used
   for congestion notification

   <S> (PDR Severe Congestion):
   1-bit.  Specifies if a severe congestion situation occurred.
   It can also carry the <S> parameter of the
   "PHR_Resource_Request" or "PHR_Refresh_Update" fields.

   <Overload %>:
   8-bit.  It includes the Overload % of the
   "PHR_Resource_Request" or "PHR_Refresh_Update" control
   information fields, indicating the level of overload to the Ingress
   node.  For more details see Section 4.6.1.6.1.

   <M> (PDR Marked):
   1-bit.  Carries the <M> value of the "PHR_Resource_Request" or
   "PHR_Refresh_Update" traffic handling directive fields.

   <B>: 1 bit Indicates bi-directional reservation.

   <Max Admitted Hops>:
   8-bit.  The <Admitted Hops> value that has been carried by the
   PHR container field used to identify the RMD reservation based node
   that admitted or process a "PHR_Resource_Request"

Bader, et al.                                                 [Page 18]


INTERNET-DRAFT                                                 RMD-QOSM

   <PDR Bandwidth>:
   32 bits.  This field specifies the bandwidth that either applies
   when the "B" flag is set to "1" and when this parameter is carried
   by a RESPONSE message, or when a severe congestion occurs and the
   QNE edges maintain an aggregated intra-domain QoS-NSLP operational
   state and it is carried by a NOTIFY message.  In the situation that
   the "B" flag is set to "1" this parameter specifies the requested
   bandwidth that have to be reserved by a node in the reverse
   direction and when the intra-domain signaling procedures require a
   bi-directional reservation procedure.  In the severe congestion
   situation this parameter specifies the bandwidth that has to be
   released.


4.2.  Message Format

   The format of the messages used by the RMD-QOSM complies with the
   QoS-NSLP and QSpec template specifications.  The QSpec used by RMD-
   QOSM is denoted in this document as RMD-QSpec and is described in
   Section 4.1.


4.3.  RMD node state management

   The QoS-NSLP state creation and management is specified in
   [QoS-NSLP].  This section describes the state creation and
   management functions of the Resource Management Function (RMF) in
   the RMD nodes.


4.3.1 Aggregated operational and reservation states at the QNE Edges

   The QNE Edges maintain both the intra-domain QoS-NSLP operational
   and reservation states, while the QNE Interior nodes maintain only
   reservation states. The structure of the intra-domain QoS-NSLP
   operational state used by the QNE edges is specified in [QoS-NSLP].

   In this case the intra-domain sessions supported by the edges are per
   aggregate sessions that have a one-to-many relationship to the per
   flow end-to-end states supported by the same edge.
   Note that the method of selecting the end-to-end sessions that form
   an aggregate is not specified in this document. An example how this
   can be accomplished is by monitoring the GIST routing states used by
   the end-to-end sessions and group the ones that use the same <PHB
   class>, QNE Ingress and QNE Egress addresses and the value of the
   priority level. Note that this priority level should be deduced from
   the priority parameters carried by the initial QSpec object.

   The operational state of this aggregated intra-domain session MUST
   contain a list with BOUND_SESSION_IDs.

   The structure of the list depends on whether a unidirectional
   reservation or a bidirectional reservation is supported.

Bader, et al.                                                 [Page 19]


INTERNET-DRAFT                                                 RMD-QOSM

   When the operational state (at QNE ingress and QNE egress) supports
   unidirectional reservations then this state MUST contain a list with
   BOUND_SESSION_IDs maintaining the SESSION_ID values of its bound
   end-to-end sessions. The BINDING_CODE associated with this
   BOUND_SESSION_ID is set to code (Aggregated sessions).  Thus the
   operational state maintains a list of BOUND_SESSION_IDs entries.
   Each entry is created when an end-to-end session joins the
   aggregated intra-domain session and is removed when an end-to-end
   session leaves the aggregate.

   It is important to emphasize that in this case, the operational
   state (at QNE ingress and QNE egress) that is maintained by each
   end-to-end session bound to the aggregated intra-domain session it
   MUST contain in the BOUND_SESSION_ID, the SESSION_ID value of the
   bound tunnelled intra-domain (aggregate) session. The BINDING_CODE
   associated with this BOUND_SESSION_ID is set to code (Aggregated
   sessions).

   When the operational state (at QNE ingress and QNE egress) supports
   bidirectional reservations then the operational state MUST contain a
   list of BOUND_SESSION_ID sets. Each set contain two
   BOUND_SESSION_IDs.  One of the BOUND_SESSION_IDs maintains the
   SESSION_ID value of one of bound end-to-end session. The
   BINDING_CODE associated with this BOUND_SESSION_ID is set to code
   (Aggregated sessions). Another BOUND_SESSION_ID, within the same set
   entry, maintains the SESSION_ID of the bidirectional bound end-to-
   end session. The BINDING_CODE associated with this BOUND_SESSION_ID
   is set to code (Bi-directional sessions).

   Note that in each set, a one to one relation exists between each
   BOUND_SESSION_ID with BINDING_CODE set to (Aggregate sessions) and
   each BOUND_SESSION_ID with BINDING_CODE set to (Bi-directional
   sessions). Each set is created when an end-to-end session joins the
   aggregated operational state and is removed when an end-to-end
   session leaves the aggregated operational state.

   It is important to emphasize that in this case, the operational
   state (at QNE ingress and QNE egress) that is maintained by each
   end-to-end session bound to the aggregated intra-domain session it
   MUST contain two types of BOUND_SESSION_IDs. One is the
   BOUND_SESSION_ID that MUST contain the SESSION_ID value of the bound
   tunelled aggregated intra-domain session that is using the
   BINDING_CODE set to (Aggregated sessions). The other
   BOUND_SESSION_ID maintains the SESSION_ID of the bound bidirectional
   end-to-end session. The BINDING_CODE associated with this
   BOUND_SESSION_ID is set to code (Bi-directional sessions).

   When the QNE Edges use aggregated QoS-NSLP reservation states, then
   the PHB class value and the size of the aggregated
   reservation, e.g., reserved bandwidth have to be maintained.
   Note that this type of aggregation is an edge to edge aggregation
   and is similar to the aggregation type specified in [RFC3175].

Bader, et al.                                                 [Page 20]


INTERNET-DRAFT                                                 RMD-QOSM

   The size of the aggregated reservations needs to be
   greater or equal to the sum of bandwidth of the inter domain
   (end-to-end) reservations/sessions it aggregates, see e.g., Section
   1.4.4 of [RFC3175].  A policy can be used
   to maintain the amount of REQUIRED bandwidth on a given aggregated
   reservation by taking into account the sum of the underlying inter
   domain (end-to-end) reservations, while endeavouring to change
   reservation less frequently.  This MAY require a trend analysis.
   If there is a significant probability that in the next interval of
   time the current aggregated reservation is exhausted, the Ingress
   router MUST predict the necessary bandwidth and request it.  If the
   Ingress router has a significant amount of bandwidth reserved but
   has very little probability of using it, the policy MAY predict the
   amount of bandwidth REQUIRED and release the excess.  To increase or
   decrease the aggregate, the RMD modification procedures SHOULD be
   used (see Section 4.6.1.4).

   The QNE interior node are reduced state nodes, i.e., they do not
   store NTLP/GIST states but they do store per PHB-aggregated QoS-NSLP
   reservation states. These reservation states are maintained and
   refreshed in the same way as described in Section 4.3.3.


4.3.2  Measurement-based method

   The QNE Edges maintain per flow intra-domain QoS-NSLP operational
   and reservation states that are containing similar data structures
   as described in Section 4.3.1. The main difference is associated
   with the different types of the used MRI and the bound end-to-end
   sessions. The structure of the maintained BOUND_SESSION_IDs depends
   on whether a unidirectional reservation or a bidirectional
   reservation is supported.

   When unidirectional reservations are supported then the operational
   state associated with this per flow intra-domain session MUST
   contain in the BOUND_SESSION_ID the SESSION_ID value of its bound
   end-to-end session. The BINDING_CODE associated with this
   BOUND_SESSION_ID is set to code (Tunnelled and end-to-end sessions).

   It is important to emphasize that in this case, the operational
   state (at QNE ingress and QNE egress) that is maintained by the end-
   to-end session bound to the per-flow intra-domain session it MUST
   contain in the BOUND_SESSION_ID, the SESSION_ID value of the bound
   tunnelled per-flow intra-domain session. The BINDING_CODE associated
   with this BOUND_SESSION_ID is set to code (Tunnelled and end-to-end
   sessions).

Bader, et al.                                                 [Page 21]


INTERNET-DRAFT                                                 RMD-QOSM

   When bidirectional reservations are supported then the operational
   state (at QNE ingress and QNE egress) MUST contain two types of
   BOUND_SESSION_IDs. One is the BOUND_SESSION_ID that maintains the
   SESSION_ID value of the bound tunnelled per-flow intra-domain
   session.  The BINDING_CODE associated with this BOUND_SESSION_ID is
   set to code (Tunnelled and end-to-end sessions).
   The other BOUND_SESSION_ID maintains the SESSION_ID of the bound
   bidirectional end-to-end session. The BINDING_CODE associated with
   this BOUND_SESSION_ID is set to code (Bi-directional sessions).

   It is important to emphasize, in this case, that the operational
   state (at ingress and egress) that is maintained by the end-to-end
   session bound to the per-flow intra-domain session MUST contain two
   types of BOUND_SESSION_IDs.  One of the BOUND_SESSION_IDs MUST
   contain the SESSION_ID of its bound end-to-end session that is using
   a BINDING_CODE with value set to code (Tunnelled and end-to-end
   sessions).  Another BOUND_SESSION_ID maintains the SESSION_ID of the
   bound bidirectional end-to-end session. The BINDING_CODE associated
   with this BOUND_SESSION_ID is set to code (Bi-directional sessions).

   Furthermore, the QoS-NSLP reservation state maintains the PHB class
   value, the value of the bandwidth requested by the end-to-end
   session bound to the intra-domain session and the value of the
   priority level.

   The measurement-based method can be classified in two schemes:

   * Congestion notification based on probing:

   In this scheme the interior nodes are Diffserv aware but not NSIS
   aware nodes. Each interior node counts the bandwidth that is used
   by each PHB traffic class. This counter value is stored in an
   RMD_QOSM state. For each traffic belonging to a PHB traffic class a
   predefined congestion threshold is set. The predefined congestion
   notification threshold is set according to, an engineered bandwidth
   limitation based on e.g. agreed Service Level Agreement or a
   capacity limitation of specific links. The threshold is usually less
   than the capacity limit, i.e., admission threshold, in order to
   avoid congestion due to the error of estimating the actual traffic
   load. The value of this threshold SHOULD be stored in another
   RMD_QOSM state.

   In this scenario end-to-end NSIS message is used as a probe packet.
   In this case the DSCP field of the GIST message is re-marked when

Bader, et al.                                                 [Page 22]


INTERNET-DRAFT                                                 RMD-QOSM

   the predefined congestion notification threshold is exceeded in an
   interior node. In this way it is ensured that the end-to-end NSIS
   message passed through the node that it is congested. This feature
   is very useful when ECMP (Equal Cost Multiple Path) based routing is
   used to detect only flows that are passing through the congested
   node. Note that in this situation, not only the probe packet is
   remarked, but also data packets passing though the congested node
   are re-marked.

   * NSIS measurement-based admission control:
   The measurement based admission control is implemented in NSIS aware
   stateless routers. Thus the main difference between this type of the
   measurement-based admission control and the congestion notification-
   based admission control is the fact that the interior nodes are NSIS
   aware nodes. In particular, the QNE Interior nodes operating
   in NSIS measurement-based mode are QoS-NSLP stateless nodes, i.e.,
   they do not support any QoS-NSLP or NTLP/GIST states.  These
   measurement-based nodes store two RMD-QOSM states per PHR group.
   These states reflect the traffic conditions at the node and are not
   affected by QoS-NSLP signaling. One state stores the measured user
   traffic load associated with the PHR group and another state stores
   the maximum traffic load threshold that can be admitted per PHR
   group. When a measurement-based node receives a intra-domain RESERVE
   message, it compares the requested resources to the available
   resources (maximum allowed minus current load) for the requested PHR
   group.  If there are insufficient resources, it sets the <M> bit in
   the RMD-QSpec.  No change to the RMD-QSpec is made when there are
   sufficient resources.


4.3.3  Reservation-based method

   The QNE Edges maintain intra-domain QoS-NSLP operational and
   reservation states that are containing similar data structures as
   described in Section 4.3.1.

   In this case the intra-domain sessions supported by the edges are per
   flow sessions that have a one-to-one relationship to the per
   flow end-to-end states supported by the same edge.
   The QNE Interior nodes operating in reservation-based mode are QoS-
   NSLP reduced state nodes, i.e., they do not store NTLP/GIST states
   but they do store per PHB-aggregated QoS-NSLP states.

   The reservation-based PHR installs and maintains one reservation
   state per PHB, in all the nodes located in the communication path.
   This state is identified by the PHB class value and it maintains the
   number of currently reserved resource units (or bandwidth).  Thus,
   the QNE Ingress node signals only the resource units requested by
   each flow.  These resource units, if admitted, are added to the
   currently reserved resources per PHB.

   For each PHB a threshold is maintained that specifies the maximum
   number of resource units that can be reserved. This threshold
   could, for example, be statically configured.

Bader, et al.                                                 [Page 23]


INTERNET-DRAFT                                                 RMD-QOSM

   An example of how the admission control and its maintenance process
   occurs in the interior nodes is described in Section 3 of [CsTa05].
   The simplified concept that is used by the per traffic class
   admission control process in the interior nodes, is based on the
   following equation:
        last + p <= T,
   where p: requested bandwidth rate, T: admission threshold, which
   reflects the maximum traffic volume that can be admitted in the
   traffic class, last: a counter that records the aggregated sum of
   the signaled bandwidth rates of previous admitted flows.

   The per-PHB group reservation states maintained in the interior
   nodes are soft states, which are refreshed by sending periodic
   refresh intra-domain RESERVE messages, which are initiated by the
   Ingress QNEs. If a refresh message corresponding to a number of
   reserved resource units (i.e., bandwidth) is not received, the
   aggregated reservation state is decreased in the next refresh period
   by the corresponding amount of resources that were not refreshed.
   The refresh period can be refined using a sliding window algorithm
   described in [RMD3].

   The reserved resources for a particular flow can also be
   explicitly released from a PHB reservation state by means of a
   intra-domain RESERVE release/tear message, which is generated by the
   Ingress QNEs.  The usage of explicit release enables the
   instantaneous release of the resources regardless of the length of
   the refresh period.  This allows a longer refresh period, which also
   reduces the number of periodic refresh messages.

   Note that both in case of measurement- and (per-flow and aggregated)
   RMD reservation-based methods,the way of how the maximum bandwidth
   thresholds are maintained is out of the specification of this
   document. However, when admission
   priorities are supported, the Maximum Allocation [RFC4125] or the
   Russian Dolls [RFC4127] bandwidth allocation model MAY be used. In
   this case three types of priority traffic classes within the same
   PHB, e.g., Expedited Forwarding, can be differentiated. These three
   different priority traffic classes, which are associated to the same
   PHB, are denoted in this document as PHB_low_priority,
   PHB_normal_priority and PHB_high_priority, and are identified by the
   PHB class value and the priority value, which is carried in the
   <Admission Priority> RMD-QSpec parameter.


4.4.  Transport of RMD-QOSM messages

   The intra-domain messages used by the RMD-QOSM SHOULD operate
   in the NTLP/GIST Datagram mode (see [GIST]).  Therefore, the NSLP
   functionality available in all QoS NSLP nodes that are able to
   support the RMD-QOSM MUST require, via the QoS-NSLP RMF API, see
   [QoS-NSLP], from the intra-domain GIST functionality available in
   these nodes to operate in the datagram mode, i.e., require GIST to:

Bader, et al.                                                 [Page 24]


INTERNET-DRAFT                                                 RMD-QOSM

   * operate in unreliable mode. This can be satisfied by passing this
     requirement from the QoS-NSLP layer to the GIST layer via the API
     Transfer-Attributes.

   * do not create a message association state. This requirement can be
     satisfied by a local policy, e.g., the QNE is configured to do not
     create a message association state

   * the interior nodes do not create any NTLP routing state.
     This can be satisfied by
     passing this requirement from the QoS-NSLP layer to the GIST layer
     via the API. However, between the QNE Egress and QNE Ingress
     routing states that are associated with intra-domain sessions
     SHOULD be created that can be used for the communication of GIST
     Data messages sent by a QNE Egress directly to a QNE Ingress. This
     type of routing state associated with an intra-domain session can
     be generated and used in the following way:

   * When the QNE Ingress has to send an initial intra-domain RESERVE
     message, the QoS-NSLP sends this message by including in the GIST
     API SendMessage primitive, the Unreliable and No security
     attributes. In order to optimise this procedure, the RMD domain
     MUST be engineered in such a way that GIST will piggyback this NSLP
     message on a GIST QUERY message. Furthermore, GIST sets the C-flag
    (C=1), see [GIST] and uses the Q-mode. The GIST functionality
     in each QNE Interior node will receive the GIST QUERY message and
     by using the RecvMessage GIST API primitive it will pass the intra-
     domain RESERVE message to the QoS-NSLP functionality. At the same
     time the GIST functionality uses the Routing-State-Check boolean
     to find out if the QoS-NSLP needs to create a routing state. The
     QoS-NSLP sets this Boolean to inform GIST to not create a routing
     state and to forward the GIST QUERY further downstream with the
     modified QoS-NSLP payload, which will include the modified intra-
     domain RESERVE message. The intra-domain RESERVE is sent in the
     same way up to the QNE Egress. The QNE Egress needs to create a
     routing state. Therefore at the moment that the GIST functionality
     passes the intra-domain RESERVE message, via the GIST RecvMessage
     primitive, to the QoS-NSLP, then at the same time the QoS-NSLP
     sets the Routing-State-Check boolean such that a routing state is
     created. The GIST creates the routing state using normal GIST
     procedures. After this phase the QNE Ingress and QNE Egress have,
     for the particular session, routing states that can route traffic
     directly from QNE Ingress to QNE Egress and from QNE Egress to
     QNE Ingress. The routing state at the QNE Egress can be used by
     the QoS-NSLP and GIST to send an intra-domain RESPONSE or intra-
     domain NOTIFY directly to the QNE Ingress using GIST Data
     messages. Note that this routing state is refreshed using normal
     GIST procedures. Note that in the above description it is
     considered that the QNE Ingress can piggyback the initial RESERVE

 Bader, et al.                                                 [Page 25]


INTERNET-DRAFT                                                 RMD-QOSM

     (NSLP) message on the GIST QUERY message. If the piggybacking of
     this NSLP (initial RESERVE) message would not be possible on the
     GIST QUERY message, then the GIST QUERY message sent by the QNE
     Ingress node would not contain any NSLP data. This GIST QUERY
     message would only be processed by the QNE Egress to generate a
     routing state. After the QNE Ingress is informed that the routing
     state at the QNE Egress is initiated, it would have to send the
     initial RESERVE message using similar procedures as for the
     situation that it would send an intra-domain RESERVE message that
     is not an initial RESERVE, see next bullet. This procedure is not
     efficient and therefore it is RECOMMENDED that the RMD domain MUST
     be engineered in such a way that the GIST protocol layer, which is
     processed on a QNE Ingress, will piggyback an initial RESERVE
    (NSLP) message on a GIST QUERY message that uses the Q-mode.

   * When the QNE Ingress needs to send an intra-domain RESERVE
     message that is not an initial RESERVE, then the QoS-NSLP sends
     this message by including in the GIST API SendMessage primitive
     the Unreliable and No security attributes. Furthermore the Local
     policy attribute is set such that GIST sends the intra-domain
     RESERVE message in a Q-mode even if there is a routing state at
     the QNE Ingress. In this way the GIST functionality uses its local
     policy to send the intra-domain RESERVE message by piggybacking it
     on a GIST DATA message and sending it in Q-mode even if there is a
     routing state for this session. The intra-domain RESERVE message
     is piggybacked on the GIST DATA message that is forwarded and
     processed by the QNE Interior nodes up to the QNE Egress.

   The transport of the original (end-to-end) RESERVE message is
   accomplished in the following way:
   At the QNE ingress the original (end-to-end) RESERVE message is
   forwarded but ignored by the stateless or reduced-state nodes, see
   Figure 3. The intermediate (interior) nodes are bypassed using
   multiple levels of NSLP-ID values (see [QoS-NSLP]). This is
   accomplished by marking the end-to-end RESERVE message, i.e.,
   modifying the QoS-NSLP default NSLP-ID value to another NSLP-ID
   predefined value.

   The marking MUST be accomplished by the ingress by modifying the
   QoS_NSLP default NSLP-ID value to a NSLP-ID predefined value. In
   This way the egress MUST stop this marking process by reassigning
   the QoS-NSLP default NSLP-ID value to the original (end-to-end)
   RESERVE message. Note that the assignment of these NSLP-ID values is
   a QOS-NSLP issue, which SHOULD be accomplished via IANA [QoS-NSLP].

Bader, et al.                                                 [Page 26]


INTERNET-DRAFT                                                 RMD-QOSM

4.5  Edge discovery and message addressing

   Mainly, the Egress node discovery can be performed either by using
   the GIST discovery mechanism [GIST], manual configuration or any
   other discovery technique.  The addressing of signaling messages
   depends on the used GIST transport mode.  The RMD-QOSM/QoS-NSLP
   signaling messages that are processed only by the Edge nodes use the
   peer-peer addressing of the GIST connection (C) mode.  RMD-QOSM/QoS-
   NSLP signaling messages that are processed by all nodes of the
   Diffserv domain, i.e., Edges and Interior nodes, use the end-end
   addressing of the GIST datagram (D) mode. Note that the RMD-QOSM
   cannot directly specify that the GIST connection or the GIST datagram
   mode SHOULD be used. This can only be specified by using, via the
   QoS-NSLP-RMF API, the GIST API Transfer-Attributes, such as
   reliable or unreliable, high or low level of
   security and by the use of local policies. RMD QoS signaling
   messages that are addressed to the data path end nodes are
   intercepted by the Egress nodes. In particular, at the ingress and
   for downstream intra-domain messages, the RMD-QOSM instructs the
   GIST functionality, via the GIST API to use among others:

   * unreliable and low level security Transfer-Attributes
   * do not create a GIST routing state
   * uses the D-mode MRI

   The intra-domain RESERVE messages can then be transported by using
   the Query D-mode, see Section 4.4..

   At the QNE Egress and for upstream intra-domain messages, the RMD-
   QOSM instructs the GIST functionality, via the GIST API to use among
   others:

   *  unreliable and low level of security Transfer-Attributes

   *  The GIST functionality uses the routing state associated with the
      intra-domain session to send an upstream intra-domain message
      directly to the QNE Ingress, see Section 4.4.


4.6.  Operation and sequence of events

4.6.1.  Basic unidirectional operation

   This section describes the basic unidirectional operation and
   sequence of events/triggers of the RMD-QOSM.  The following basic
   operation cases are distinguished:

Bader, et al.                                                 [Page 27]


INTERNET-DRAFT                                                 RMD-QOSM

   * Successful reservation (Section 4.6.1.1),
   * Unsuccessful reservation (Section 4.6.1.2),
   * RMD refresh reservation (Section 4.6.1.3),
   * RMD modification of aggregated reservation (4.6.1.4)
   * RMD release procedure (Section 4.6.1.5.)
   * Severe congestion handling (Section 4.6.1.6.)
   * Admission control using congestion notification based on probing
     (Section 4.6.1.7.).

   The QNEs at the Edges of the RMD domain support the RMD QoS Model and
   end-to-end QoS models, which process the RESERVE message differently.
   Note that the term end-to-end QoS model applies to any QoS model that
   is initiated and terminated outside the RMD-QOSM aware domain.
   However, there might be situations where a QoS model is initiated
   and/or terminated by the QNE Edges and is considered to be an end-to-
   end QoS model. This can occur when the QNE Edges can also operate as
   either QNI or as QNR and at the same time they can operate as either
   sender or receiver of the data path. Note that the described
   functionality described in Sections 4.6.1.1, 4.6.1.2, 4.6.1.3,
   4.6.1.5, and 4.6.1.6 applies to the RMD reservation-based and to
   the NSIS measurement-basedadmission control methods.
   The described functionality in Section 4.6.1.7
   applies to the admission control procedure that uses the congestion
   notification based on probing. The QNE Edge nodes maintain either per
   flow QoS-NSLP operational and reservation states or aggregated QoS-
   NSLP operational and reservation states.

   When the QNE Edges maintain aggregated QoS-NSLP operational and
   reservation states, the RMD-QOSM functionality MAY accomplish a RMD
   modification procedure (see Section 4.6.1.4), instead of the
   reservation initiation procedure that is described in this
   subsection. Note that it is RECOMMENDED that the QNE implementations
   of RMD-QOSM process the QoS-NSLP signaling messages with a higher
   priority than data packets. This can be accomplished as described in
   Section 3.3.4 of [QoS-NSLP] and it can be requested via the QoS-NSLP-
   RMF API described in [QoS-NSLP]. The signalling scenarios described
   in this section are accomplished using the QoS-NSLP processing rules
   defined in [QoS-NSLP], in combination with the RMF triggers sent via
   the QoS-NSLP-RMF API described in [QoS-NSLP].


4.6.1.1.  Successful reservation

   This section describes the operation of the RMD-QOSM where a
   reservation is successfully accomplished.

   The QNI generates the initial RESERVE message, and it is forwarded
   by the NTLP as usual [GIST].

Bader, et al.                                                 [Page 28]


INTERNET-DRAFT                                                 RMD-QOSM

4.6.1.1.1. Operation in Ingress node

   When an end-to-end reservation request (RESERVE) arrives at the
   Ingress node (QNE), see Figure 8, it is processed based on the end-
   to-end QoS model.   Subsequently, the RMD QoS Description:
   <Bandwidth>, <PHB Class>, <Admission Priority> are derived from the
   <QoS Desired> object of the initial QSpec.

   The value of the <Bandwidth> parameter used by the RMD-Qspec is found
   by copying the value of the "Peak Data Rate [p]" of the <TMOD-1>
   parameter into the <Bandwidth> parameter.

   If the initial QSpec does not contain the <PHB Class> parameter,
   then the selection of the <PHB class> that is carried by the intra-
   domain RMD-QSpec is defined by a local policy similar to the
   procedures discussed in [RFC2998] and [RFC3175].
   For example, in the situation that the initial QSpec is used by
   the IntServ Controlled Load QOSM then the Expedited Forwarding (EF)
   PHB is appropriate to set the <PHB class> parameter carried by the
   intra-domain RMD-QSpec, see [RFC3175].

   If the initial QSpec does not carry the <Admission Priority>
   parameter then the <Admission Priority> parameter in the RMD-QSpec
   will not be populated. If the initial QSpec does not carry the
   <Admission Priority> parameter, but it carries other priority
   parameters, then it is considered that edges as being stateful nodes,
   are able to control the priority of the sessions that are
   entering or leaving the RMD domain in accordance to the priority
   parameters. Note that the RMF reservation states, see Section 4.3, in
   the QNE edges store the value of the <Admission Priority> parameter
   that is used within the RMD domain in case of pre-emption and severe
   congestion situations, see Section 4.6.1.6.
   If the RMD domain supports pre-emption during the admission control
   process, then the QNE Ingress node can support the building
   blocks specified in the [QoS-NSLP] and during the admission
   control process use the example pre-emption handling algorithm
   described in Appendix 4.
   Note that in the above described case, the QNE egress uses, if
   available, the tunnelled initial priority parameters, which can
   be interpreted by the QNE egress.

   If the initial QSpec carries the <Excess Treatment> parameter,
   then the QNE ingress and QNE egress nodes MUST control the excess
   traffic that is entering or leaving the RMD domain in accordance to
   the <Excess Treatment> parameter. Note that the RMD-QSpec does not
   carry the <Excess Treatment> parameter. However, by using the <PHB
   class> parameter the RMD domain uses the excess treatment procedures
   specified by the particular PHB standard.

Bader, et al.                                                 [Page 29]


INTERNET-DRAFT                                                 RMD-QOSM

   If the requested <TMOD-1> parameter carried by the initial QSpec,
   cannot be satisfied, then an end to end RESPONSE message has to be
   generated. However, in order to decide whether the end-to-end
   reservation request was locally (at the QNE Ingress) satisfied, also
   a local(at the QNE_Ingress) RMD-QoSM admission control procedure has
   to be performed. In other words, the RMD-QOSM functionality has to
   verify whether the value included in the RMD-QOSM <Bandwidth> can be
   reserved and stored in the RMD-QOSM reservation states, see Sections
   4.6.1.1.2 and 4.3.

   An initial QSpec object MUST be included in the end-to-end
   RESPONSE message. The parameters included in the QSpec <QoS
   Reserved> object are copied from the original <QoS Desired> values.
   The "E" flag associated with the QSPEC <QoS Reserved> object and the
   "E" flag associated with the <Bandwidth> parameter are set. In
   addition, the INFO-SPEC object is included in the end to end RESPONSE
   message. The error code used by this INFO-SPEC is:

   Error severity class: Transient Failure
   Error code value: Reservation failure

   Furthermore, all the other RESPONSE parameters are set according to
   the end-to-end QoS model or according to [QoS-NSLP] and [QSP-T].

   If the request was satisfied locally (see Section 4.3), the Ingress
   QNE node generates two RESERVE messages: one intra-domain and
   one end-to-end RESERVE message. Note however, that when the
   aggregated QOS-NSLP operational and reservation states are used by
   the QNE Ingress, then the generation of the intra-domain RESERVE
   message depends on the availability of the aggregated QoS-NSLP
   operational state. If this aggregated QoS-NSLP operational state is
   available, then the RMD modification of aggregated reservations
   described in section 4.6.1.4. is used.

   It is important to note that when "per flow RMD reservation based"
   scenario is used within the RMD domain, the retransmission within the
   RMD domain SHOULD be disallowed. The reason of this is related to the
   fact that the QNI Interior nodes are not able to differentiate
   between a retransmitted RESERVE message associated with a certain
   session and an initial RESERVE message belonging to another session.
   However, the QNE Ingress have to report a failure situation upstream.
   When the QNE Ingress transmits the (intra-domain or end-to-end)
   RESERVE with RII object set, it waits for a RESPONSE from the QNE
   Egress for a QOSNSLP_REQUEST_RETRY period.

   If the QNE Ingress transmitted an intra-domain or end-to-end RESERVE
   message with the RII object set and it fails to receive the
   associated intra-domain or end-to-end RESPONSE, respectively, after
   the QOSNSLP_REQUEST_RETRY period expires, it considers that the
   reservation failed. In this case the QNE Ingress SHOULD generate an
   end-to-end RESPONSE message that will include among others an
   INFO-SPEC object. The error code used by this INFO-SPEC is:
      Error severity class: Transient Failure
      Error code value: Reservation failure

Bader, et al.                                                 [Page 30]


INTERNET-DRAFT                                                 RMD-QOSM

   Furthermore, all the other RESPONSE parameters are set according to
   the end-to-end QoS model or according to [QoS-NSLP] and [QSP-T].

   Note however, that if the retransmission within the RMD domain is not
   disallowed then the procedure described in Appendix
   A.5 SHOULD be used on QNE Interior nodes, see also [Chan07]. In this
   case the stateful QNE Ingress uses the retransmission procedure
   described in [QoS-NSLP].

   If a rerouting takes place then the stateful QNE
   ingress is following the procedures specified in [QoS-NSLP].

   At this point the intra-domain and end-to-end operational states MUST
   be initiated or modified according to the REQUIRED binding
   procedures.

   The way of how the BOUND_SESSION_IDs are initiated and maintained in
   the intra-domain and end-to-end QoS-NSLP operational states is
   described in Section 4.3.1 and 4.3.2.

   These two messages are bound together in the following way. The end-
   to-end RESERVE SHOULD contain in the BOUND_SESSION_ID the SESSION_ID
   of its bound intra-domain session.

   Furthermore, if the QNE Edge nodes maintain intra-domain per flow
   QoS-NSLP reservation states then the value of Binding_Code MUST be
   set to code "Tunnel and end-to-end sessions", see Section 4.3.2.

   In addition to this then the intra-domain and end-to-end RESERVE
   messages are bound using the Message binding procedure described
   in [QoS-NSLP]. In particular the <MSG_ID> object is included
   in the intra-domain RESERVE message and its bound <BOUND_MSG_ID>
   object is carried by the end-to-end RESERVE message. Furthermore, the
   Message_Binding_Type flag is SET (value is 1), such that the message
   dependency is bi-directional.

   If the QOS-NSLP edges maintain aggregated intra-domain QoS-NSLP
   operational states then the value of Binding_Code MUST be set to code
   "Aggregated sessions".
   Furthermore, in this case the retransmission within the RMD domain
   SHOULD NOT be disallowed and the procedures described in Appendix
    A.5 SHOULD be used on QNE Interior nodes. This is necessary due to
   the fact that when retransmissions are disallowed then the
   associated with (micro) flows belonging to the aggregate will loose
   their reservations. Note that in this case the stateful QNE Ingress
   uses the retransmission procedure described in [QoS-NSLP].

Bader, et al.                                                 [Page 31]


INTERNET-DRAFT                                                 RMD-QOSM

   The intra-domain RESERVE message is associated with the (local NTLP)
   SESSION_ID mentioned above. The selection of the IP source and IP
   destination address of this message depends on how the
   different inter-domain (end-to-end) flows are aggregated by the
   QNE Ingress node (see Section 4.3.1). As described in Section 4.3.1,
   the QNE Edges maintain either per flow, or aggregated QoS-NSLP
   reservation states for the RMD QoS model, which are identified by
   (local NTLP) SESSION_IDs (see [GIST]). Note that this NTLP SESSION ID
   is a different one than the SESSION_ID associated with the end-to-end
   RESERVE message.

   If no QOS-NSLP aggregation procedure at the QNE Edges is supported
   then the IP source and IP destination address of this message MUST be
   equal to the IP Source and IP destination addresses of the data flow.
   The intra-domain RESERVE message is sent using the NTLP datagram
   mode (see Sections 4.4, 4.5). Note that the GIST datagram mode can be
   selected using the unreliable GIST API Transfer-Attributes. In
   addition, the intra-domain RESERVE (RMD-QSpec) message MUST include a
   PHR container (PHR_Resource_Request) and the "RMD-QOSM object
   combination".

   The end-to-end RESERVE message includes the initial QSpec and it
   is sent towards the Egress QNE.

   Note that after completing the initial discovery phase, the GIST
   connection mode can be used between the QNE Ingress and QNE Egress.
   Note that the GIST connection mode can be selected using the reliable
   GIST API Transfer-Attributes.

   The end-to-end RESERVE message is forwarded using the GIST
   forwarding procedure to bypass the Interior stateless or reduced-
   state QNE nodes, see Figure 8.  The bypassing procedure is
   described in Section 4.4. At the QNE Ingress the end-to-end RESERVE
   message is marked, i.e., modifying the QoS-NSLP default NSLP-ID value
   to another NSLP-ID predefined value that will be used by the GIST
   message carrying the end-to-end RESPONSE message to bypass the QNE
   Interior nodes. Note that the QNE Interior nodes, see [GIST], are
   configured to handle only certain NSLP-Ids, see [QoS-NSLP].

   Furthermore, note that the initial discovery phase and the process of
   sending the end-to-end RESERVE message towards the QNE Egress MAY be
   done simultaneously. This can be accomplished only if the GIST
   implementation is configured to perform that, via e.g., a local
   policy. However, the selection of the discovery procedure cannot be
   selected by the RMD-QOSM.

   The (initial) intra-domain RESERVE message MUST be sent by the QNE
   Ingress and it MUST contain the following values (see QoS-NSLP-RMF
   API described in [QoS-NSLP]):
   *  the value of the <RSN> object is generated and processed as
      described in [QoS-NSLP];

Bader, et al.                                                 [Page 32]


INTERNET-DRAFT                                                 RMD-QOSM

   *  the SCOPING flag MUST NOT be set, meaning that a default
      scoping of the message is used.  Therefore, the QNE Edges MUST
      be configured as RMD boundary nodes and the QNE Interior nodes
      MUST be configured as Interior (intermediary) nodes;

   *  the <RII> MUST be included in this message, see [QoS-NSLP].

   *  The flag REPLACE MUST be set to FALSE = 0;

*  The value of the Message ID value carried by the <MSG_ID> object
   is set according to [QoS-NSLP]. The value of the
   Message_binding_Type is set to "1".

*  the value of the <REFRESH_PERIOD> object MUST be calculated
   and set by the QNE Ingress node as described in Section 4.6.1.3;

*  the value of the <PACKET_CLASSIFIER> object is associated with
   the path-coupled routing MRM, since RMD-QOSM is used with the
   path-coupled MRM. The flag that has to be set is
   the flag T (traffic class) meaning that the packet classification
   of packets is based on the DSCP value included in the IP header of
   the packets. Note that the DSCP value used in the MRI can be
   derived by the value of <PHB class> parameter, which MUST be
   carried by the intra-domain RESERVE message. Note that the QNE
   Ingress being a QNI for the intra-domain session it can pass this
   value to GIST, via the GIST API.

*  the PHR resource units MUST be included into the <Bandwidth>
   parameter of the "<QoS Desired> object. When the QNE edges
   use per flow intra-domain QoS-NSLP states, then the value of the
   <Bandwdith> parameter can be obtained by using the method of
   copying the peak rate (p) field included in the <TMOD-1> parameter
      carried by the initial QSpec into this <Bandwidth> parameter,
      which is described above in this subsection. When the QNE edges
      use aggregated intra-domain QoS-NSLP operational states, then the
      value of the <Bandwdith> parameter can be obtained by using the
      bandwidth aggregation method described in Section 4.3.1;

   *  the value of the <PHB class> parameter can be defined by using the
      method of copying the <PHB Class> parameter carried by
      the initial QSpec into the <PHB class> carried by the RMD-QSpec,
      which is described above in this subsection.

   *  the value of the Parameter/Container ID field of the PHR container
      MUST be set to PHR_1, (i.e., PHR_Resource_Request;)

   *  the value of the <Admitted Hops> parameter in the PHR container
      MUST be set to "1". Note that during a successful reservation each
      time a RMD-QOSM aware node processes the RMD-QSpec, the <Admitted
      Hops> parameter is increased by one.

   *  the value of the <Hop_U> parameter in the PHR container MUST be
      set to "0";

Bader, et al.                                                 [Page 33]


INTERNET-DRAFT                                                 RMD-QOSM

   *  the value of the <Max Admitted Hops> is set to "0".

   * If the initial QSpec carried an <Admission Priority>
     parameter, then this parameter SHOULD be copied into the RMD-QSpec
     and carried by the
     (initiating) intra-domain RESERVE. Note that for the RMD-QOSM a
     reservation established without an <Admission Priority> parameter
     is equivalent to a reservation with <Admission Priority> value 1.
     Note that in this case each admission priority is associated with
     a priority traffic class. The three priority traffic classes
     (PHB_low_priority, PHB_normal_priority, PHB_high_priority) MAY be
     associated with the same PHB, see Section 4.3.3.

   *  In a single RMD domain case the PDR container MAY not be included
      into the message.

   Note that the intra-domain RESERVE message does not carry the
   BOUND_SESSION_ID object. The reason of this is that the end-to-end
   RESERVE carries in the BOUND_SESSION_ID object the SESSION_ID value
   of the intra-domain session.

   When an end-to-end RESPONSE message is received by the QNE
   Ingress node, which was sent by a QNE Egress node see Section
   4.6.1.1.3, then it is processed according to [QoS-NSLP]
   and end-to-end QoS model rules.

   When an intra-domain RESPONSE message is received
   by the QNE Ingress node, which was sent by a QNE Egress see Section
   4.6.1.1.3, it uses the QoS-NSLP procedures to match it to the earlier
   sent intra-domain RESERVE message. After this phase, the RMD-QSpec
   has to be identified and processed.

   The RMD QoSM reservation has been successful if the <M> bit carried
   by the "PDR Container" is equal to "0" (i.e., not set).

   Furthermore, the INFO_SPEC object is processed as defined in the
   QoS-NSLP specification. In case of successful reservation the
   INFO_SPEC object MUST have the following values:

   * Error Severity Class: Success
   * Error Code value: Reservation successful

   If the end-to-end RESPONSE message has to be forwarded to a
   node outside the RMD-QOSM aware domain then the values of
   the objects contained in this message (i.e., <RII/RSN>, <INFO_SPEC>,
   [QSPEC ]) MUST be set by the QOS-NSLP protocol functions
   of the QNE. If an end-to-end QUERY is received by the QNE Ingress
   then the same bypassing procedure has to be used as the one applied
   for an end-to-end RESERVE message. In particular, it is forwarded
   using the GIST forwarding procedure to bypass the Interior stateless
   or reduced-state QNE nodes.


Bader, et al.                                                 [Page 34]


INTERNET-DRAFT                                                 RMD-QOSM

4.6.1.1.2 Operation in the Interior nodes

   Each QNE Interior node MUST use the QoS-NSLP and RMD-QOSM parameters
   of the intra-domain RESERVE (RMD-QSpec) message as follows (see
   QoS-NSLP-RMF API described in [QoS-NSLP]):

  *   the values of the <RSN>, <RII>, <PACKET_CLASSIFIER>,
      <REFRESH_PERIOD>, objects MUST NOT be changed.
      The interior node is informed by the <PACKET_CLASSIFIER> object
      that the packet classification SHOULD be done on the DSCP value.
      The flag that has to be set in this case is the flag T (traffic
      class). The value of the DSCP value MUST
      be obtained via the MRI parameters that the QoS-NSLP receives from
      GIST. A QNE Interior MUST be able to associate the value carried
      by the RMD-QSpec <PHB class> parameter and the DSCP value obtained
      via GIST. This is REQUIRED, because there are situations that the
      <PHB class> parameter is not carrying a DSCP value, but a "PHB ID
      code", see Section 4.1.1.

   *  The flag REPLACE MUST be set to FALSE = 0;

   *  when the RMD reservation based methods described in Section 4.3.1
      and 4.3.3 are used, the value of <Bandwidth> parameter of the "RMD
      QoS Description" field is used by the QNE Interior node for
      admission control. Furthermore, if the <Admission Priority>
      parameter is carried by the  <QoS Desired> object, then this
      parameter is processed as described in the following bullets.

   *  in case of the RMD reservation-based procedure, and if these
      resources are admitted (see Section 4.3.1, 4.3.3), they are added
      to the currently reserved resources. Furthermore, the value of the
      <Admitted Hops> parameter in the PHR container has to be increased
      by one.

   *  If the bandwidth allocated for the PHB_high_priority traffic is
      fully utilized, and a high priority request arrives, other
      policies on allocating bandwidth can be used, which are beyond the
      scope of this document.

   *  If the RMD domain supports pre-emption during the admission
      control process, then the QNE Interior node can support the
      building blocks specified in the [QoS-NSLP] and during the
      admission control process use the pre-emption handling algorithm
      specified in Appendix 4.

   *  in case of the RMD measurement based method (see Section 4.3.2),
      and if the requested value of the <Bandwidth> parameter is
      admitted, using a MBAC algorithm, then the number of this
      resources will be used to update the MBAC algorithm according to
      the operation described in Section 4.3.2.

Bader, et al.                                                 [Page 35]


INTERNET-DRAFT                                                 RMD-QOSM


4.6.1.1.3 Operation in the Egress node

   When the end-to-end RESERVE message is received by the egress node,
   it is only forwarded further, towards QNR, if the processing of the
   intra-domain RESERVE(RMD-QSpec) message was successful at all nodes
   in the RMD domain. In this case, the QNE Egress MUST stop the marking
   process that was used to bypass the QNE Interior nodes by reassigning
   the QoS-NSLP default NSLP-ID value to the end-to-end RESERVE message,
   see Section 4.4. Furthermore, the carried BOUND_SESSION_ID object
   associated with the intra-domain session MUST be removed after
   processing. Note that the received end to end RESERVE was tunneled
   within the RMD domain. Therefore, the tunnelled initial QSpec
   carried by the end-to-end RESERVE message has to be processed/set
   according to the [QSP-T] specification.

   If a rerouting takes place, then the stateful QNE egress is following
   the procedures specified in [QoS-NSLP].
   At this point the intra-domain and end-to-end operational states MUST
   be initiated or modified according to the REQUIRED binding
   procedures.

   The way of how the BOUND_SESSION_IDs are initiated and maintained in
   the intra-domain and end-to-end QoS-NSLP operational states is
   described in Section 4.3.1 and 4.3.2.

   If the processing of the intra-domain RESERVE(RMD-QSpec) was not
   successful at all nodes in the RMD domain then the inter-domain (end-
   to-end) reservation is considered as being failed. Furthermore, note
   that when the QNE Egress uses per flow intra-domain QoS-NSLP
   operational states, see Sections 4.3.2 and 4.3.3, the QNE Egress
   SHOULD support the message binding procedure described in [QoS-NSLP],
   which can be used to synchronize the arrival of the end to end
   RESERVE and the intra-domain RESERVE (RMD-QSpec) messages, see
   Section 5.7 and QoS-NSLP-RMF API described in [QoS-NSLP]. Note that
   the intra-domain RESERVE message carries the <MSG_ID> object and its
   bound end-to-end RESERVE message carries the <BOUND_MSG_ID> object.
   Both these objects carry the Message_Binding_Type flag set to the
   value of 1. If these two messages do not arrive during the time
   defined by the MsgIDWait timer, then the reservation is considered as
   being failed. Note that the timer has to be pre-configured and it has
   to have the same value in the RMD domain. In this case an end-to-end
   RESPONSE message, see QoS-NSLP-RMF API described in [QoS-NSLP], is
   sent towards the QNE ingress with the following INFO_SPEC values:

   Error Class: Transient Failure
   Error Code: Mismatch synchronization between end-to-end RESERVE
   and intra-domain RESERVE

Bader, et al.                                                 [Page 36]


INTERNET-DRAFT                                                 RMD-QOSM

   When the intra-domain RESERVE(RMD-QSpec) is received by the QNE
   Egress node of the session associated with the intra-domain
   RESERVE(RMD-QSpec) (the PHB session) with the session included in
   its <BOUND_SESSION_ID> object MUST be bound according to the
   specification given in [QoS-NSLP].  The SESSION_ID included
   in the BOUND_SESSION_ID parameter stored in the intra-domain QoS-NSLP
   operational state object is the SESSION_ID of the session associated
   with the end-to-end RESERVE message(s). Note that if the QNE Edge
   nodes maintain per flow intra-domain QoS NSLP operational states then
   the value of Binding_Code = (Tunnel and end-to-end sessions) is used
   If the QNE Edge nodes maintain per aggregated QoS-NSLP intra-domain
   reservation states then the value of Binding_Code = (Aggregated
   sessions), see Sections 4.3.1, 4.3.2.

   If the RMD domain supports pre-emption during the admission control
   process, then the QNE Egress node can support the building
   blocks specified in the [QoS-NSLP] and during the admission
   control process use the example pre-emption handling algorithm
   described in Appendix 4.
   The end-to-end RESERVE message is generated/forwarded further
   upstream according to the [QoS-NSLP] and [QSP-T] specifications.

   Furthermore, the "B" (BREAK) QoS-NSLP flag in the end to end
   RESERVE message MUST NOT be set, see QoS-NSLP-RMF API described in
   QoS-NSLP.

QNE (Ingress)     QNE (Interior)        QNE (Interior)    QNE (Egress)
NTLP stateful    NTLP stateless        NTLP stateless    NTLP stateful
    |                    |                   |                    |
RESERVE                  |                   |                    |
--->|                    |                   |     RESERVE        |
    |------------------------------------------------------------>|
    |RESERVE(RMD-QSpec)  |                   |                    |
    |------------------->|                   |                    |
    |                    |RESERVE(RMD-QSpec) |                    |
    |                    |------------------>|                    |
    |                    |                   | RESERVE(RMD-QSpec) |
    |                    |                   |------------------->|
    |                    |RESPONSE(RMD-QSpec)|                    |
    |<------------------------------------------------------------|
    |                    |                   |                RESERVE
    |                    |                   |                    |-->
    |                    |                   |                RESPONSE
    |                    |                   |                    |<--
    |                    |RESPONSE           |                    |
    |<------------------------------------------------------------|
RESPONSE                 |                   |                    |
<---|                    |                   |                    |

Figure 8: Basic operation of successful reservation procedure used by
          the RMD-QOSM

Bader, et al.                                                 [Page 37]


INTERNET-DRAFT                                                 RMD-QOSM

   The QNE Egress MUST generate an intra-domain RESPONSE (RMD-Qspec)
   message. The intra-domain RESPONSE (RMD-QSpec) message MUST
   be sent to the QNE Ingress node, i.e., the previous stateful hop by
   using the procedures described in Sections 4.4 and 4.5.
   The values of the RMD-QSpec that is carried by the intra-domain
   RESPONSE message MUST be used and/or set in the following way
   (see QoS-NSLP-RMF API described in [QoS-NSLP]):

   *  the RII object carried by the intra-domain RESERVE message, see
      Section 4.6.1.1.1, has to be copied and carried by the
      intra-domain RESPONSE message.

   *  the value of the Parameter/Container ID field of the PDR container
      MUST be set "PDR_7" (i.e., PDR_Reservation_Report);

   *  the value of the <M> field of the PDR container MUST be equal to
      the value of the <M> parameter of the PHR container that was
      carried by its associated intra-domain RESERVE(RMD-QSpec)
      message. This is REQUIRED since the value of the <M> parameter is
      used to indicate the status if the RMD reservation request to the
      Ingress edge.

   If the binding between the intra-domain session and the end-to-end
   session uses a Binding_Code is (Aggregated sessions), and there is no
   aggregated QoS-NSLP operational state associated with the intra-
   domain session available, then the RMD
   modification of aggregated reservation procedure described in Section
   4.6.1.4. can be used.

   If the QNE Egress receives an end-to-end RESPONSE message, it is
   processed and forwarded towards the QNE Ingress. In particular, the
   non-default values of the objects contained in the end-to-end
   RESPONSE message MUST be used and/or set by the QNE Egress as
   follows (see QoS-NSLP-RMF API described in [QoS-NSLP]):

   * the values of the <RII/RSN>, <INFO_SPEC>, [ QSPEC ] objects are
     set according to [QoS-NSLP] and/or [QSP-T]. The INFO_SPEC object
     SHOULD be set by the QoS-NSLP functionality. In case of successful
     reservation the INFO_SPEC object SHOULD have the following values:
     Error Severity Class: Success,
     Error Code value: Reservation successful,

   * Furthermore, an initial QSpec object MUST be included in the
     end-to-end RESPONSE message. The parameters included in the QSPEC
    <QoS Reserved> object are copied from the original <QoS Desired>
     values.

   The end-to-end RESPONSE message are delivered as normal, i.e.,
   is addressed and sent to its upstream QoS-NSLP neighbor, i.e., QNE
   Ingress node.

Bader, et al.                                                 [Page 38]


INTERNET-DRAFT                                                 RMD-QOSM

   Note that if a QNE Egress receives an end-to-end QUERY that was
   bypassed through the RMD domain, it MUST stop the marking
   process that was used to bypass the QNE Interior nodes. This can be
   done by reassigning the QoS-NSLP default NSLP-ID value to the end-to-
   end QUERY message, see Section 4.4.


4.6.1.2.  Unsuccessful reservation

   This section describes the operation where a request for reservation
   cannot be satisfied by the RMD-QOSM.

   The QNE Ingress, the QNE Interior and QNE Egress nodes process and
   forward the end-to-end RESERVE message and the intra-domain
   RESERVE(RMD-QSpec) message in a similar way as specified in Section
   4.6.1.1.  The main difference between the unsuccessful operation and
   successful operation is that one of the QNE nodes does not admit the
   request due to lack of resources.  This also means that the QNE edge
   node MUST NOT forward the end-to-end RESERVE message towards the
   QNR node.

   Note that the described functionality applies to the RMD reservation-
   based methods, see Sections 4.3.1, 4.3.2, and to the NSIS
   measurement-based admission control method, see Section 4.3.2.
   The QNE Edge nodes maintain either per flow QoS-NSLP reservation
   states or aggregated QoS-NSLP reservation states. When the QNE edges
   maintain aggregated QoS-NSLP reservation states, the RMD-QOSM
   functionality MAY accomplish a RMD modification procedure (see
   Section 4.6.1.4.), instead of the reservation initiation procedure
   that is described in this subsection.


4.6.1.2.1 Operation in the Ingress nodes

   When an end-to-end RESERVE message arrives at the QNE Ingress and
   if there are no resources available, the QNE Ingress MUST
   reject this end-to-end RESERVE message and send an end-to-end
   RESPONSE message back to the sender, as described in the QoS-NSLP
   specification, see [QoS-NSLP] and [QSP-T].

   When an end-to-end RESPONSE message is received by an Ingress
   node, see Section 4.6.1.2.3, the values of the <RII/RSN>,
   [<INFO_SPEC> ], [<QSPEC>] objects are processed according to the QoS-
   NSLP procedures.

   If the end-to-end RESPONSE message has to be forwarded upstream to a
   node outside the RMD-QOSM aware domain then the values of
   the objects contained in this message (i.e., <RII/RSN>, <INFO_SPEC>,
   [ QSPEC ]) MUST be set by the QOS-NSLP protocol functions of the QNE.

Bader, et al.                                                 [Page 39]


INTERNET-DRAFT                                                 RMD-QOSM

   When an intra-domain RESPONSE message is received by the QNE Ingress
   node, which was sent by a QNE Egress, see Section 4.6.1.2.3, it uses
   the QoS-NSLP procedures to match it to the earlier sent intra-domain
   RESERVE message. After this phase, the RMD-QSpec has to be identified
   and processed. Note that in this case the RMD Resource Management
   Function (RMF) is notified that the reservation has been
   unsuccessful, by reading the <M> parameter of the PDR container.
   Note that when the QNE edges maintain a per flow QoS-NSLP reservation
   state the RMD-QOSM functionality, has to start an RMD release
   procedure (see Section 4.6.1.5). When the QNE edges maintain
   aggregated QoS-NSLP reservation states the RMD-QOSM functionality MAY
   start a RMD modification procedures (see Section 4.6.1.4.).


4.6.1.2.2 Operation in the Interior nodes

   In case of the RMD reservation based scenario, and if the
   intra-domain reservation request is not admitted by the QNE Interior
   node then the <Hop_U> and <M> parameters of the PHR container MUST be
   set to "1".  The <Admitted Hops> counter MUST NOT be increased.
   Moreover, the value of the <Max Admitted Hops> MUST be set equal to
   the <Admitted Hops> value.
   Furthermore, the "E" flag associated with the QSpec <QoS Desired>
   object and the "E" flag associated with the <Bandwidth> parameter
   SHOULD be set. In case of the RMD measurement based scenario, the
   <M> parameter of the PHR container MUST be set to "1". Furthermore,
   the "E" flag associated with the QSpec <QoS Desired> object and the
   "E" flag associated with the <Bandwidth> parameter SHOULD be set.
   Note that the <M> flag seems to be set in a similar way as the "E"
   flag used by the <Bandwidth> parameter. However, the ways of how the
   two flags are processed by a QNE are different.

   In general, if a QNE Interior node receives a QSpec <Bandwidth>
   parameter with the "E" flag set and a PHR container type
   "PHR_Resource_Request", with the <M> parameter set to "1" , then this
   "PHR Container" and the "RMD-QoSM object combination" (i.e., RMD-QOSM
   <QoS Desired> object) MUST NOT be processed. Furthermore, when the
   <K> parameter that is included in the "PHR Container" and carried by
   a RESERVE message is set to "1", then this "PHR Container" and the
   "RMD QoS Description" (i.e., RMD-QOSM <QoS Desired> object) MUST NOT
   be processed.


4.6.1.2.3 Operation in the Egress nodes

   In the RMD reservation based, see Sections 4.3.3, and the RMD NSIS
   measurement based scenario, see Section 4.3.2, when the <M> marked
   intra-domain RESERVE(RMD-QSpec) is received by the QNE Egress node
   (see Figure 9) the session associated with the intra-domain
   RESERVE(RMD-QSpec) (the PHB session) and the end-to-end session MUST
   be bound.

Bader, et al.                                                 [Page 40]


INTERNET-DRAFT                                                 RMD-QOSM

   When the QNE Egress uses per flow intra-domain QoS-NSLP operational
   states, see Section 4.3.2 and 4.3.3, then the QNE Egress node MUST
   generate an end-to-end RESPONSE message that has to be sent to its
   previous stateful QoS-NSLP hop (see QoS-NSLP-RMF API described in
   [QoS-NSLP]).

   *  the values of the <RII/RSN>, <INFO_SPEC> objects are set
      by the standard QoS-NSLP protocol functions. In case of the
      unsuccessful reservation the INFO_SPEC object SHOULD have the
      following values:
      Error Severity Class: Transient Failure
      Error Code value: Reservation failure

   The QSpec that was carried by the end to end RESERVE belonging to the
   same session as this end-to-end RESPONSE is included in this message.

   In particular, the parameters included in the QSpec <QoS Reserved>
   object of the end-to-end RESPONSE message are copied
   from the initial <QoS Desired> values included in its associated
   end-to-end RESERVE message. The "E" flag associated with
   the QSpec <QoS Reserved> object and the "E" flag associated with the
   <TMOD-1> parameter included in the end-to-end RESPONSE are set.

QNE (Ingress)    QNE (Interior)       QNE (Interior)      QNE (Egress)
NTLP stateful    NTLP stateless        NTLP stateless    NTLP stateful
    |                    |                   |                    |
RESERVE                  |                   |                    |
--->|                    |                   |     RESERVE        |
    |------------------------------------------------------------>|
    |RESERVE(RMD-QSpec:M=0)                  |                    |
    |------------------->|                   |                    |
    |                    |RESERVE(RMD-QSpec:M =1)                 |
    |                    |------------------>|                    |
    |                    |                   | RESERVE(RMD-QSpec:M=1)
    |                    |                   |------------------->|
    |                    |RESPONSE(RMD-QOSM) |                    |
    |<------------------------------------------------------------|
    |                    |RESPONSE           |                    |
    |<------------------------------------------------------------|
RESPONSE                 |                   |                    |
<---|                    |                   |                    |
RESERVE(RMD-QSpec: Tear=1, M=1, <Admitted Hops>=<Max Admitted Hops>
    |------------------->|                   |                    |
                         |RESERVE(RMD-QSpec: Tear=1, M=1, K=1)    |
    |                    |------------------>|                    |
                         |    RESERVE(RMD-QSpec: Tear=1, M=1, K=1)|
    |                    |                   |------------------->|

Figure 9: Basic operation during unsuccessful reservation
          initiation used by the RMD-QOSM

Bader, et al.                                                 [Page 41]


INTERNET-DRAFT                                                 RMD-QOSM

   In addition to the above, similarly to the successful operation,
   see Section 4.6.1.1.3, the QNE Egress MUST generate an intra-domain
   RESPONSE message that has to be sent to its previous stateful QoS-
   NSLP hop. The values of the <RII/RSN>, <INFO_SPEC> objects are set by
   the standard QoS-NSLP protocol functions. In case of the unsuccessful
   reservation the INFO_SPEC object SHOULD have the following values
   (see QoS-NSLP-RMF API described in [QoS-NSLP]):

      Error Severity Class: Transient Failure
      Error Code value: Reservation failure

   The values of the RMD-QSpec MUST be used and/or set
   in the following way (see QoS-NSLP-RMF API described in [QoS-NSLP]):

   *  the value of the <PDR Control Type> of the PDR container MUST be
      set to "PDR_7" (PDR_Reservation_Report);

   *  the value of the <Max Admitted Hops> parameter of the PHR
       container included in the received <M> marked intra-domain
      RESERVE (RMD-QSpec)  MUST be included
      in the <Max Admitted Hops> parameter of the PDR container;

   *  the value of the <M> parameter of the PDR container MUST be "1".


4.6.1.3 RMD refresh reservation

   In case of RMD measurement-based method, see Section 4.3.2, QoS-NSLP
   reservation states in the RMD domain are typically not maintained,
   therefore, this method typically does not use an intra-domain refresh
   procedure. However, there are measurement based optimization schemes,
   see [GrTs03], which MAY use the refresh procedures described in
   Sections 4.6.1.3.1, and 4.6.1.3.3. However, this measurement based
   optimization schemes can only be applied in the RMD domain if the QNE
   edges are configured to perform intra-domain refresh procedures and
   if all the QNE interior nodes are configured to perform the
   measurement based optimization schemes. In the description given in
   this subsection it is assumed that the RMD measurement based scheme
   does not use the refresh procedures.

   When the QNE edges maintain aggregated or per flow QoS-NSLP
   operational and reservation states, see Sections 4.3.1 and 4.3.3,
   then the refresh procedures are very similar. If the RESERVE messages
   arrive within the soft state time-out period, the corresponding
   number of resource units are not removed. However, the transmission
   of the intra-domain and end-to-end (refresh) RESERVE message are not
   necessarily synchronized. Furthermore, the generation of the end-to-
   end RESERVE message, by the QNE edges, depends on the locally
   maintained refreshed interval (see [QoS-NSLP]).

Bader, et al.                                                 [Page 42]


INTERNET-DRAFT                                                 RMD-QOSM


4.6.1.3.1 Operation in the Ingress node

   The Ingress node MUST be able to generate an intra-domain (refresh)
   RESERVE(RMD-QSpec) at any time defined by the refresh period/timer.
   Before generating this message, the RMD QoS signaling model
   functionality is using the RMD traffic class (PHR) resource units for
   refreshing the RMD traffic class state.

   Note that the RMD traffic class refresh periods MUST be equal in
   all QNE edge and QNE Interior nodes and SHOULD be smaller (default:
   more than two times smaller) than the refresh period at the QNE
   Ingress node used by the end-to-end RESERVE message. The intra-domain
   RESERVE (RMD-QSpec) message MUST include a "RMD-QOSM object
   combination" (i.e., RMD-QOSM <QoS Desired>) and a PHR container
   (i.e., PHR_Refresh_Update).

   An example of this refresh operation can be seen in Figure 10.

QNE (Ingress)    QNE (Interior)        QNE (Interior)    QNE (Egress)
NTLP stateful    NTLP stateless        NTLP stateless    NTLP stateful
    |                    |                   |                    |
    |RESERVE(RMD-QSpec)  |                   |                    |
    |------------------->|                   |                    |
    |                    |RESERVE(RMD-QSpec) |                    |
    |                    |------------------>|                    |
    |                    |                   | RESERVE(RMD-QSpec) |
    |                    |                   |------------------->|
    |                    |                   |                    |
    |                    |RESPONSE(RMD-QSpec)|                    |
    |<------------------------------------------------------------|
    |                    |                   |                    |

   Figure 10: Basic operation of RMD specific refresh procedure

   Most of the non-default values of the objects contained in this
   message MUST be used and set by the QNE Ingress in the same
   way as described in Section 4.6.1.1.  The following objects are
   used and/or set differently:

  *  the PHR resource units MUST be included into the <Bandwidth>
     parameter. The value of the <Bandwidth> parameter depends on
     how the different inter domain (end-to-end) flows are aggregated
     by the QNE Ingress node (e.g., the sum of all the PHR requested
     resources of the aggregated flows), see Section 4.3.1.  If no
     QOS-NSLP aggregation is accomplished by the QNE Ingress node, the
     value of the <Bandwidth> parameter SHOULD be equal to the
     <Bandwidth> parameter of its associated new (initial) intra-domain
     RESERVE (RMD-QSpec) message, see Section 4.3.3. ;

Bader, et al.                                                 [Page 43]


INTERNET-DRAFT                                                 RMD-QOSM

   *  the value of the Parameter/Container field of the "PHR Container"
       MUST be set to "PHR_2",
      i.e., "PHR_Refresh_Update";

   When the intra-domain RESPONSE (RMD-QSpec) message, see Section
   4.6.1.3.3., is received by the QNE Ingress node, then:

   *  the values of the <RII/RSN>, <INFO_SPEC>, [QSP-T] objects are
      processed by the standard QoS-NSLP protocol functions (see Section
      4.6.1.1.);

   *  the "PDR Container" has to be processed by the RMD-QOSM
      functionality in the QNE Ingress node.  The RMD-QOSM functionality
      is notified by the <PDR M> parameter of the PDR container
      that the refresh procedure has been successful or unsuccessful.
      All session(s) (when aggregated QoS-NSLP operational and
      reservation states are used, see Section 4.3.1, there
      will be more than one sessions) associated with this RMD specific
      refresh session MUST be informed about the success or failure of
      the refresh procedure.  In case of failure, the QNE Ingress node
      has to generate (in a standard QoS-NSLP way) an error end-to-end
      RESPONSE message that will be sent towards QNI.


4.6.1.3.2 Operation in the Interior node

   The intra-domain RESERVE (RMD-QSpec) message is received and
   processed by the QNE Interior nodes.  Any QNE edge or QNE Interior
   node that receives a "PHR_Refresh_Update" field
   MUST identify the traffic class state (PHB) (using the
   <PHB Class> parameter).  Most of the parameters in this refresh
   intra-domain RESERVE (RMD-QSpec) message MUST be used and/or set by
   a QNE Interior node in the same way as described in Section 4.6.1.1.

   The following objects are used and/or set differently:

   * the value of <Bandwidth> parameter of the "RMD-QOSM object
     combination" (i.e., the RMD-QOSM <QoS Desired>)
     is used by the QNE Interior node for refreshing the RMD
     traffic class state. These resources (included in <Bandwidth>),
     if reserved, are added to the currently reserved resources
     per PHB and therefore they will become a part of the per traffic
     class (per-PHB) reservation state, see Sections 4.3.1 and 4.3.3. If
     the refresh procedure cannot be fulfilled then the <M> and <S>
     fields carried by the PHR container MUST be set to "1".

   * Furthermore, the "E" flag associated with <QoS Desired> object and
     the "E" flag associated with the <Bandwidth> parameter SHOULD be
     set.

Bader, et al.                                                 [Page 44]


INTERNET-DRAFT                                                 RMD-QOSM

   Any PHR container of type "PHR_Refresh_Update", and its associated
   "RMD-QOSM object combination" (i.e., <Bandwidth>), whether it is
   marked or not and independent of the "E" flag value of the
   <Bandwdith> parameter, is always processed, but marked bits are not
   changed.


4.6.1.3.3 Operation in the Egress node

   The intra-domain RESERVE(RMD-QSpec) message is received and
   processed by the QNE Egress node.  A new intra-domain RESPONSE
   (RMD-QSpec) message is generated by the QNE Egress node and MUST
   include a PDR (type PDR_Refresh_Report).

   The (refresh) intra-domain RESPONSE (RMD-QSpec) message MUST be sent
   to the QNE Ingress node, i.e., the previous stateful hop. The
   (refresh) intra-domain RESPONSE (RMD-QSpec) message MUST be
   explicitly routed to the QNE Ingress node, i.e., the previous
   stateful hop, using the procedures described in Section 4.5.

   *  the values of the <RII/RSN>, <INFO_SPEC> objects are set
     by the standard QoS-NSLP protocol functions, see [QoS-NSLP].

   * The value of the <PDR Control Type> parameter of the PDR container
     MUST be set "PDR_8" (i.e. PDR_Refresh_Report).
     In case of successful reservation the INFO_SPEC object SHOULD
     have the following values:
     Error Severity Class: Success
     Error Code value: Reservation successful

   * In case of unsuccessful reservation the INFO_SPEC object SHOULD
     have the following values:
     Error Severity Class: Transient Failure
     Error Code value: Reservation failure

   The RMD-QSpec that was carried by the intra-domain RESERVE
   belonging to the same session as this intra-domain RESPONSE is
   included in the intra-domain RESPONSE message. The parameters
   included in the QSPec <QoS Reserved> object are copied from the
   original <QoS Desired> values. If the reservation is unsuccessful
   then "E" flag associated with the QSpec <QoS Reserved> object and the
   "E" flag associated with the <Bandwidth> parameter are set.
    Furthermore, the <M> and <S> PDR Container bits are set to "1".


4.6.1.4.  RMD modification of aggregated reservations

   In the case when the QNE edges maintain QoS-NSLP aggregated
   operational and reservation states and the aggregated reservation has
   to be modified (see Section 4.3.1) the following procedure is
   applied:

Bader, et al.                                                 [Page 45]


INTERNET-DRAFT                                                 RMD-QOSM

   * When the modification request requires an increase of the reserved
     resources, the QNE Ingress node MUST include the corresponding
     value into the <Bandwidth> parameter of the "RMD-QOSM object
     combination", (i.e., the RMD-QOSM <QoS Desired>)
     which is sent together with a "PHR_Resource_Request" control
     information.  If a QNE edge or QNE Interior node is not able to
     reserve the number of requested resources, the
     "PHR_Resource_Request" that is associated with
     the <Bandwidth> parameter MUST be <M> marked, i.e., the <M> bit is
     set to the value of "1".  In this situation the
     RMD specific operation for unsuccessful reservation will be applied
     (see Section 4.6.1.2).

   * When the modification request requires a decrease of the
     reserved resources, the QNE Ingress node MUST include this value
     into the <Bandwidth> parameter of the "RMD-QOSM object combination"
     (i.e., the RMD-QOSM <QoS Desired>). Subsequently an RMD release
     procedure SHOULD be accomplished (see Section 4.6.1.5). Note that
     if the complete bandwidth associated with the aggregated
     reservation maintained at the QNE ingress does not have to be
     released then the TEAR flag MUST be set to OFF. This is because the
     NSLP operational states associated with the aggregated reservation
     states at the edge QNEs MUST NOT be turned off. However, if the
     complete bandwidth associated with the aggregated reservation
     maintained at the QNE ingress has to be released, then the TEAR
     flag MUST be set to ON.


4.6.1.5  RMD release procedure

   This procedure is applied to all RMD mechanisms that maintain
   reservation states. If a refresh RESERVE message does not arrive at a
   QNE Interior node within the refresh time-out period then the
   bandwidth requested by this refresh RESERVE message is not updated.
   This will mean that the reserved bandwidth associated with the
   reduced state is decreased in the next refresh period by the amount
   of the corresponding bandwidth that has not been refreshed, see
   section 4.3.3. This soft state
   behavior provides certain robustness for the system ensuring that
   unused resources are not reserved for long time.  Resources can be
   removed by an explicit release at any time. However, in the situation
   that an end-to-end (tear) RESERVE is retransmitted, see Section 5.2.4
   in [QoS-NSLP], then this message MUST NOT initiate an intra-domain
   (tear) RESERVE message. This is because the amount bandwidth within
    the RMD domain associated with the (tear) end-to-end RESERVE has
    already been released and therefore, this amount of bandwidth within
    the RMD domain MUST NOT once again be released.

Bader, et al.                                                 [Page 46]


INTERNET-DRAFT                                                 RMD-QOSM

   When the RMD-RMF of a QNE edge or QNE Interior node processes a
   "PHR_Release_Request"  PHR container it MUST identify the
   <PHB Class> parameter and estimate the time period that elapsed
   after the previous refresh, see also Section 3 of [CsTa05]. This MAY
   be done by indicating the time lag, say "T_lag", between the last
   sent "PHR_Refresh_Update" and the "PHR_Release_Request" control
   information container by the QNE Ingress node, see [RMD1] and
   [CsTa05] for more details.  The value of "T_Lag"
   is first normalized to the length of the refresh period, say
   "T_period".  The ratio between the "T_Lag" and the length of the
   refresh period, "T_period", is calculated.  This ratio is then
   introduced into the <Time Lag> field of the "PHR_Release_Request".

   When a node (QNE Egress or QNE Interior) receives the
   "PHR_Release_Request" PHR container, it MUST store the arrival
   time.  Then it MUST calculate the time difference, "Tdiff", between
   the arrival time and the start of the current refresh period,
   "T_period".  Furthermore, this node MUST derive the value of the
   "T_Lag", from the <Time Lag> parameter. "T_Lag" can be found by
   multiplying the value included in the <Time Lag> parameter with the
   length of the refresh period, "T_period". If the derived time lag,
   "T_lag", is smaller than the calculated time difference, "T_diff",
   then this node MUST decrease the PHB reservation state with the
   number of resource units indicated in the <Bandwidth> parameter of
   the "RMD-QOSM object combination" (i.e., the RMD-QOSM <QoS Desired>)
   that has been sent together with the "PHR_Release_Request" "PHR
   Container", but not below zero.

   An RMD specific release procedure can be triggered by an end-to-end
   RESERVE with a TEAR flag set ON (see Section 4.6.1.5.1) or it can be
   triggered by either an intra-domain RESPONSE, an end-to-end RESPONSE
    or an end-to-end NOTIFY message that includes a marked (i.e., PDR
   <M> and/or PDR <S> parameters are set ON) "PDR_Reservation_Report" or
   "PDR_Congestion_Report" and/or an INFO_SPEC object.


4.6.1.5.1.  Triggered by a RESERVE message

   This RMD explicit release procedure can be triggered by a tear (TEAR
   flag set ON) end-to-end RESERVE message. When a tear (TEAR flag
   set ON) end-to-end RESERVE message arrives to the QNE Ingress
   then the QNE Ingress node SHOULD process the message in a standard
   QoS-NSLP way (see [QoS-NSLP]). In addition to this, the RMD RMF
   is notified, as specified in [QoS-NSLP].

   Same as for the scenario described in Section 4.6.1.1., a bypassing
   procedure has to be initiated by the QNE Ingress node. The bypassing
   procedure is performed according to the description given in Section
   4.4. At the QNE Ingress the end-to-end RESERVE message is marked,
   i.e., modifying the QoS-NSLP default NSLP-ID value to another NSLP-ID
   predefined value that will be used by the GIST message that carries
   the end-to-end RESERVE message to bypass the QNE Interior nodes.

Bader, et al.                                                 [Page 47]


INTERNET-DRAFT                                                 RMD-QOSM

   Before generating an intra-domain tear RESERVE, the RMD-QOSM has to
   release the requested RMD-QOSM bandwidth from the RMD traffic class
   state maintained at the QNE Ingress. This can be achieved by
   identifying the traffic class (PHB) and then subtracting the amount
   of RMD traffic class requested resources, included in the <Bandwidth>
   parameter, from the total reserved amount of resources stored in the
   RMD traffic class state. The <Time Lag> is used as explained in the
   introductory part of Section 4.6.1.5.

QNE (Ingress)     QNE (Interior)       QNE (Interior)    QNE (Egress)
NTLP stateful    NTLP stateless        NTLP stateless    NTLP stateful
    |                    |                   |                    |
RESERVE                  |                   |                    |
--->|                    |                   |     RESERVE        |
    |------------------------------------------------------------>|
    |RESERVE(RMD-QSpec:Tear=1)               |                    |
    |------------------->|                   |                    |
    |                    |RESERVE(RMD-QSpec:Tear=1)               |
    |                    |------------------->|                   |
    |                    |                 RESERVE(RMD-QSpec:Tear=1)
    |                    |                   |------------------->|
    |                    |                   |                RESERVE
    |                    |                   |                    |-->

   Figure 11: Explicit release triggered by RESERVE used by the RMD-QOSM

   After that the REQUIRED bandwidth is released from the RMD-QOSM
   traffic class state at the QNE Ingress, an intra-domain RESERVE (RMD-
   QOSM) message has to be generated. The intra-domain RESERVE (RMD-
   QSpec) message MUST include a "RMD QoS object combination" field and
   a PHR container, (i.e., "PHR_Resource_Release") and it MAY include a
   PDR container, (i.e., PDR_Release_Request). An example of this
   operation can be seen in Figure 11.

   Most of the non default values of the objects contained in the
   tear intra-domain RESERVE message are set by the QNE Ingress node in
   the same way as described in Section 4.6.1.1.  The following objects
   are set differently (see QoS-NSLP-RMF API described in [QoS-NSLP]):

   *  The <RII> object MUST NOT be included in this message.  This is
      because the QNE Ingress node does not need to receive a
      response from the QNE Egress node;

   *  If the release procedure is not applied for the RMD modification
      of aggregated reservation procedure, see Section 4.6.1.4, then
      the TEAR flag MUST be set to ON;

   *  the PHR resource units MUST be included into the <Bandwidth>
      parameter of the "RMD-QOSM object combination" (i.e., the RMD-QOSM
      <QoS Desired>);

   *  the value of the <Admitted Hops> parameter MUST be set to "1";

Bader, et al.                                                 [Page 48]


INTERNET-DRAFT                                                 RMD-QOSM

   *  the value of the <Time Lag> parameter of the PHR container is
      calculated by the RMD-QOSM functionality (see 4.6.1.5) the value
      of the <Control Type> parameter of PHR container is set to "PHR_3"
      (i.e., PHR_Resource_Release).

   Any QNE Interior node that receives the combination of the RMD-
   QOSM <Qos Desired> object and the "PHR_Resource_Release" control
   information container  MUST identify the traffic class (PHB)
   and release the requested resources included in the <Bandwidth>
   parameter.  This can be achieved by subtracting the amount of RMD
   traffic class requested resources, included in the <Bandwidth>
   parameter, from the total reserved amount of resources stored in the
   RMD traffic class state.  The value of the <Time Lag> parameter of
   the "PHR_Resource_Release" container is used during the release
   procedure as explained in the introductory part of Section 4.6.1.5.

   The intra-domain tear RESERVE (RMD-QSpec) message is received and
   processed by the QNE Egress node.  The "RMD-QOSM object combination"
   (i.e., the RMD-QOSM <QoS Desired>) and the "PHR RMD-QOSM control"
   container (and if available the "PDR Container") are read and
   processed by the RMD QoS node.

   The value of the <Bandwidth> parameter of the "RMD-QOSM object
   combination" (i.e., the RMD-QOSM <QoS Desired>)
   and the value of the <Time Lag> field of the PHR container MUST
   be used by the RMD release procedure.  This can be achieved by
   subtracting the amount of RMD traffic class requested resources,
   included in the <Bandwidth> parameter, from the total reserved amount
   of resources stored in the RMD traffic class state.

   The end-to-end RESERVE message is forwarded by the next hop (i.e.,
   the QNE Egress) only if the intra-domain tear RESERVE (RMD-QSpec)
   message arrives at the QNE Egress node. Furthermore, the QNE Egress
   MUST stop the marking process that was used to bypass the QNE
   Interior nodes by reassigning the QoS-NSLP default NSLP-ID value to
   the end-to-end RESERVE message, see Section 4.4.
   Note that when the QNE edges maintain aggregated QoS-NSLP reservation
   states the RMD-QOSM functionality MAY start a RMD modification
   procedures (see Section 4.6.1.4.) that uses the explicit release
   procedure described above, in this subsection. Note that
   if the complete bandwidth associated with the aggregated reservation
   maintained at the QNE ingress has to be released then the
   TEAR flag MUST be set to ON. Otherwise, the TEAR
   flag MUST be set to OFF, see Section 4.6.1.4.


Bader, et al.                                                 [Page 49]


INTERNET-DRAFT                                                 RMD-QOSM

4.6.1.5.2   Triggered by a marked RESPONSE or NOTIFY message

   This RMD explicit release procedure can be triggered by either an
   intra-domain RESPONSE message with a PDR container carrying among
   others the <M> and <S> parameters with values <M>=1 and <S>=0 (see
   Section 4.6.1.2) an intra-domain (refresh) RESPONSE message carrying
   a PDR Container with <M>=1 and <S>=1  (see Section 4.6.1.6.1) or an
   end to end NOTIFY message (see Section 4.6.1.6.) with an INFO_SPEC
   object with the following values:

   Error Severity Class: Informational
   Error Code value: Congestion situation

   When the aggregated intra-domain QoS-NSLP operational states are used
   then an end-to-end NOTIFY message used to trigger an RMD release
   procedure MAY contain a PDR container that carries a <M> and a <S>
   with values <M>=1 and <S>=1, and a bandwidth value in the <PDR
   Bandwidth> parameter included in a "PDR_Refresh_Report" or
   "PDR_Congestion_Report" container.

   Note that in all explicit release procedures, before generating an
   intra-domain tear RESERVE, the RMD-QOSM has to release the requested
   RMD-QOSM bandwidth from the RMD traffic class state maintained at the
   QNE Ingress. This can be achieved by identifying the traffic class
   (PHB) and then subtracting the amount of RMD traffic class requested
   resources, included in the <Bandwidth> parameter, from the total
   reserved amount of resources stored in the RMD traffic class state.

   Figure 12 shows the situation that the intra-domain tear RESERVE is
   generated after being triggered by either an intra-domain (refresh)
   RESPONSE message that carries a PDR Container with <M>=1 and <S>=1,
   or by an end-to-end NOTIFY message that do not carry a PDR container,
   but an INFO_SPEC object. The error code values carried by this NOTIFY
   message are:

   Error Severity Class: Informational
   Error Code value: Congestion situation

   Most of the non-default values of the objects contained in the
   tear intra-domain RESERVE(RMD-QSpec) message are set by the QNE
   Ingress node in the same way as described in Section 4.6.1.1.

   The following objects MUST be used and/or set differently (see
   QoS-NSLP-RMF described in [QoS-NSLP]):

   *  The value of the <M> parameter of the PHR container MUST be set
      to "1".

   *  the value of the <S> parameter of the
      "PHR container" MUST be set to "1".

Bader, et al.                                                 [Page 50]


INTERNET-DRAFT                                                 RMD-QOSM

   *  The RESERVE message MAY include a PDR container. Note that this
      is needed if bi-directional scenario is used, see Section 4.6.2.

QNE (Ingress)     QNE (Interior)         QNE (Interior)    QNE (Egress)
NTLP stateful    NTLP stateless         NTLP stateless    NTLP stateful
    |                  |                  |                  |
    | NOTIFY           |                  |                  |
    |<-------------------------------------------------------|
    |RESERVE(RMD-QSpec:Tear=1,M=1,S=1)    |                  |
    | ---------------->|RESERVE(RMD-QSpec:Tear=1, M=1,S=1)   |
    |                  |                  |                  |
    |                  |----------------->|                  |
    |                  |           RESERVE(RMD-QSpec:Tear=1, M=1,S=1)
    |                  |                  |----------------->|

   Figure 12: Basic operation during RMD explicit release procedure
   triggered by NOTIFY used by the RMD-QOSM.

   Note that if the values of the <M> and <S> parameters included in the
   PHR container carried by a intra-domain tear RESERVE(RMD-QOSM) are
   set as ((<M>=0 and <S>=1) or (<M>=0 and <S>=0) or (<M>=1 and <S>=1))
   then the <Max Admitted Hops> value SHOULD NOT be compared to the
   <Admitted Hops> value and the value of the <K> field MUST NOT be set.
   Any QNE edge or QNE Interior node that receives the intra-domain tear
   RESERVE it MUST check the <K> field included in the PHR Container. If
   the <K> fied is "0" then the traffic class state (PHB) has to be
   identified, using the <PHB Class> parameter, and the the requested
   resources included in the <Bandwidth> field have to be released.
   This can be achieved by subtracting the amount of RMD traffic class
   requested resources, included in the <Bandwidth> field, from the
   total reserved amount of resources stored in the RMD traffic class
   state.  The value of the <Time Lag> parameter of the PHR field
   is used during the release procedure as explained in the introductory
   part of Section 4.6.1.5. Afterwards, the QNE Egress node MUST
   terminate the tear intra-domain RESERVE(RMD-QSpec) message.

   The RMD specific release procedure that is triggered by an
   intra-domain RESPONSE message with <M>=1 and <S>=0 PDR container (see
   Section 4.6.1.2) generates an intra-domain tear RESERVE message
   that uses the combination of <Max Admitted Hops> and <Admitted_Hops>
   fields to calculate and specify when the <K> value carried by the
   "PHR Container" can be set. When the <K> field is set, then the "PHR
   Container" and the "RMD-QoSM object combination" carried by an
   intra-domain tear RESERVE MUST NOT be processed.

   The RMD specific explicit release procedure that uses the combination
   of <Max Admitted Hops>, <Admitted_Hops> and <K> fields to release
   resources/bandwidth in only a part of the RMD domain, is denoted as
   RMD partial release procedure.  This explicit release procedure can
   be used, for example, during unsuccessful reservation (see Section
   4.6.1.2). When the RMD-QoSM/QoS-NSLP signaling model functionality
   of a QNE Ingress node receives a  PDR container with values <M>=1
   and <S>=0, of type "PDR_Reservation_Report", it MUST start an RMD
   partial release procedure.

 Bader, et al.                                                 [Page 51]


INTERNET-DRAFT                                                 RMD-QOSM

   In this situation, after that the REQUIRED bandwidth is released from
   the RMD-QOSM traffic class state at the QNE Ingress, an intra-domain
   RESERVE (RMD-QOSM) message has to be generated. An example of this
   operation can be seen in Figure 13.

   Most of the non-default values of the objects contained in the
   tear intra-domain RESERVE(RMD-QSpec) message are set by the QNE
   Ingress node in the same way as described in Section 4.6.1.1.

   The following objects MUST be used and/or set differently:


   *  The value of the <M> parameter of the PHR container MUST be set
      to "1".

   *  The RESERVE message MAY include a PDR container.

* the value of the <Max Admitted Hops> carried by the "PHR
     Container" MUST be set equal to the <Max Admitted Hops> value
     carried by the "PDR Container" (with <M>=1 and <S.=0) carried by
     the received intra-domain RESPONSE message that triggers the
     release procedure.

QNE (Ingress)     QNE (Interior)        QNE (Interior)    QNE (Egress)
                                     Node that marked
                                    PHR_Resource_Request
                                       <PHR> object
NTLP stateful    NTLP stateless        NTLP stateless    NTLP stateful
    |                    |                   |                    |
    |                    |                   |                    |
    | RESPONSE (RMD-QSpec: M=1)              |                    |
    |<------------------------------------------------------------|
RESERVE(RMD-QSpec: Tear=1, M=1, <Admit Hops>=<Max Admitted Hops>, K=0)
    |------------------->|                   |                    |
    |                    |RESERVE(RMD-QSpec: Tear=1, M=1, K=1)    |
    |                    |------------------>|                    |
    |                    |    RESERVE(RMD-QSpec: Tear=1, M=1, K=1)|
    |                    |                   |------------------->|
    |                    |                   |                    |

   Figure 13: Basic operation during RMD explicit release procedure
   Triggered by RESPONSE used by the RMD-QOSM

   Any QNE edge or QNE Interior node that receives the intra-domain tear
   RESERVE has to check the value of the <K> field in the "PHR
   Container" before releasing the requested resources.

   If the value of the <K> field is "1", then all the QNEs located
   downstream, including the QNE Egress, MUST NOT process the carried
   "PHR Container" and the "RMD-QOSM Object Combination" (i.e., RMD-QOSM
   <QoS Desired> object) by the intra-domain tearing RESERVE.

Bader, et al.                                                 [Page 52]


INTERNET-DRAFT                                                 RMD-QOSM

   If the <K> field value is "0", any QNE edge or QNE Interior node that
   receives the intra-domain tear RESERVE can release the resources by
   subtracting the amount of RMD traffic class requested resources,
   included in the <Bandwidth> field, from the total reserved amount of
   resources stored in the RMD traffic class state.  The value of the
   <Time Lag> parameter of the PHR field is used during the release
   procedure as explained in the introductory part of Section 4.6.1.5.
   Furthermore, the QNE MUST perform the following procedures.
   If the values of the <M> and <S> parameters included in the
    "PHR_Resource_Release" PHR container are (<M=1> and <S>=0) then the
   <Max Admitted Hops> value MUST be compared with the calculated
   <Admitted Hops> value.  Note that each time that the intra-domain
   tear RESERVE is processed and before being forwarded by a QNE, the
   <Admitted Hops> value included in the PHR container is increased by
   one.

   When these two values are equal then the intra-domain
   RESERVE(RMD-QSpec) that is forwarded further towards the QNE Egress
   MUST set the <K> value of the carried "PHR Container" to "1".
   The reason of doing this is that the QNE node that is currently
   processing this message was the last QNE node that successfully
   processed the "RMD-QOSM object combination" (i.e., the RMD-QOSM <QoS
   Desired>) and PHR container of its associated initial reservation
   request (i.e., initial intra-domain RESERVE(RMD-QSpec) message). Its
   next QNE downstream node was unable to successfully process the
   initial reservation request, therefore, this QNE node marked the <M>
   and <Hop_U> parameters of the "PHR_Resource_Request".

   Note that finally the QNE Egress node MUST terminate the intra-domain
   RESERVE(RMD-QSpec) message.

   Note that the above described RMD partial release procedure applies
   to the situation that the QNE edges maintain a per flow QoS-NSLP
   reservation state.

   When the QNE edges maintain aggregated intra-domain QoS-NSLP
   operational states and a severe congestion occurs, then the QNE
   Ingress MAY receive an end to end NOTIFY message (see Section
   4.6.1.6.) with a PDR container that carries the  <M>=0 and <S>=1
   fields and a bandwidth value in the <PDR Bandwidth> parameter
   included in a "PDR_Congestion_Report" container. Furthermore the
   same end-to-end NOTIFY message carries an INFO_SPEC object with the
   following values:

   Error Severity Class: Informational
   Error Code value: Congestion situation

 Bader, et al.                                                 [Page 53]


INTERNET-DRAFT                                                 RMD-QOSM

   The end-to-end session associated with this NOTIFY message maintains
   the BOUND_SESSION_ID of the bound aggregated session, see Sections
   4.3.1. The RMD-QOSM at QNE Ingress MUST start a RMD modification
   procedures (see Section 4.6.1.4) that uses the RMD explicit release
   procedure described above in this section. In particular, the RMD
   explicit release procedure releases the bandwidth value included in
   the <PDR Bandwidth> parameter, within the "PDR_Congestion_Report"
   container, from the reserved bandwidth associated with the aggregated
   intra-domain QoS-NSLP operational state.

4.6.1.6. Severe congestion handling

   This section describes the operation of the RMD-QOSM when a severe
   congestion occurs within the Diffserv domain.

   When a failure in a communication path, e.g., a router or a link
   failure occurs, the routing algorithms will adapt to failures by
   changing the routing decisions to reflect changes in the topology and
   traffic volume.  As a result, the re-routed traffic will follow a new
   path, which MAY result in overloaded nodes as they need to support
   more traffic.  This MAY cause severe congestion in the communication
   path.  In this situation the available resources, are not enough to
   meet the REQUIRED QoS for all the flows along the new path.
   Therefore, one or more flows SHOULD be terminated, or forwarded in a
   lower priority queue.

   Interior nodes notify edge nodes by data marking or marking the
   refresh messages.

4.6.1.6.1 Severe congestion handling by the RMD-QOSM refresh procedure

   This procedure applies to all RMD scenarios that use a RMD refresh
   procedure. The QoS-NSLP and RMD are able to cope with congested
   situations using the refresh procedure, see Section 4.6.1.3. If the
   refresh is not successful in an QNE Interior node, edge nodes are
   notified by setting <S>=1 (<M>=1) marking the refresh messages and by
   including the percentage of overload into the <Overload %> field in
   the "PHR_Refresh_Update" container, carried by the intra-domain
   RESERVE message. The intra-domain RESPONSE message that is sent by
   the QNE Egress towards QNE Ingress will contain a PDR container with
   a Parameter/Container ID = PDR_10, i.e., "PDR_Congestion_Report". The
   values of the <M>, <S> and <Overload %> fields of this container
   SHOULD be set equal to the values of the <M>, <S> and <Overload %>
   fields, respectively, carried by the "PHR_Refresh_Update" container.
   Part of the flows, corresponding to the <Overload %>, are terminated,
   or forwarded in a lower priority queue. Note that an example of how
   this value can be calculated is given in appendix A.1.1 and denoted
   as the signaled_overload_rate parameter. The flows can be terminated
   by the RMD release procedure described in Section 4.6.1.5.  Note that
   the above described functionality applies to the RMD reservation-
   based and to the NSIS measurement-based admission control schemes.
   Furthermore, note that the above functionalities apply also for the
   scenario where the QNE Edge nodes maintain either per flow QoS-NSLP
   reservation states or aggregated QoS-NSLP reservation states.

 Bader, et al.                                                 [Page 54]


INTERNET-DRAFT                                                 RMD-QOSM

   In general, relying on the soft state refresh mechanism solves the
   congestion within the time frame of the refresh period. If this
   mechanism is not fast enough additional functions SHOULD be used,
   which are described in Section 4.6.1.6.2.


4.6.1.6.2 Severe congestion handling by proportional data packet marking

   This severe congestion handling method requires the following
   functionalities.


4.6.1.6.2.1 Operation in the Interior nodes

   The detection and marking/remarking functionality described in this
   section applies to NSIS aware, but also to NSIS unaware nodes. This
   means however, that the "not NSIS aware" nodes MUST be configured
   such that they can detect the congestion/severe congestion situations
   and remark packets in the same way as the "NSIS aware" nodes do.

   The Interior node detecting severe congestion remarks data packets
   passing the node. For this remarking, two additional DSCPs can be
   allocated for each traffic class.  One DSCP MAY be used to indicate
   that the packet passed a congested node. This type of DSCP is denoted
   in this document as "affected DSCP" and is used to indicate that a
   packet passed through a severe congested node. The use of this DSCP
   type eliminates the possibility that, due to e.g. ECMP (Equal Cost
   Multiple Paths) enabled routing, the egress node either does not
   detect packets passed a severe congested node or erroneously detects
   packets that actually did not pass the severe congested node.  Note
   that this type of DSCP MUST only be used if all the nodes within the
   RMD domain are configured to use it. Otherwise, this type of DSCP
   MUST NOT be applied. The other DSCP MUST be used to indicate the
   degree of congestion by marking the bytes proportionally to the
   degree of congestion. This type of DSCP is denoted in this document
   as "encoded DSCP".

   Note that in this document the terms marked packets or marked bytes
   refer to the "encoded DSCP". The terms unmarked packets or unmarked
   bytes are representing the packets or the bytes belonging to these
   packets that their DSCP is either the "affected DSCP" or the original
   DSCP. Furthermore, in the algorithm described below it is considered
   that the router MAY drop received packets. The counting/measuring of
   marked or unmarked bytes described in this section is accomplished
   within measurement periods. All nodes within a RMD domain use the
   same, fixed measurement interval, say T seconds, which MUST be
   pre-configured.

 Bader, et al.                                                 [Page 55]


INTERNET-DRAFT                                                 RMD-QOSM

   It is RECOMMENDED that the total number of additional (local and
   experimental) DSCPs needed for severe congestion handling within an
   RMD domain SHOULD be as low as possible and it SHOULD NOT exceed the
   limit of 8. One possibility to reduce the number of used DSCPs is to
   use only the "encoded DSCP" and not to use "affected DSCP" marking.
   Another possible solution is for example, to allocate one DSCP for
   severe congestion indication for each of the AF classes that can be
   supported by RMD-QOSM.

   An example of a remarking procedure can be found in Appendix A.1.1.


4.6.1.6.2.2 Operation in the Egress nodes

   When the QNE edges maintain a per flow intra-domain QoS-NSLP
   operational state, see sections 4.3.2, 4.3.3, then the following
   procedure is followed. The QNE Egress node applies a predefined
   policy to solve the severe congestion situation, by selecting a
   number of inter-domain (end-to-end) flows that SHOULD be terminated,
   or forwarded in a lower priority queue.

   When the RMD domain does not use the "affected DSCP"
   marking then the egress MUST generate an ingress/egress pair
   aggregated state, for each ingress and for each supported PHB. This
   is because the edges MUST be able to detect in which ingress/egress
   pair a severe congestion occurs. This is because otherwise the QNE
   Egress will not have any information on which flows or groups of
   flows were affected by the severe congestion. When the RMD domain
   supports the "affected DSCP" marking then the egress is able to
   detect all flows that are affected by the severe congestion
   situation. Therefore, when the RMD domain supports the "affected
   DSCP" marking, then the Egress MAY not generate and maintain the
   ingress/egress pair aggregated reservation States. Note that these
   aggregated reservation states MAY not be associated with aggregated
   intra-domain QoS-NSLP operational states.

   The ingress/egress pair aggregated reservation state can be derived
   by detecting, which flows are using the same PHB and are sent by the
   same Ingress (via the per flow end-to-end QoS-NSLP states).

   Some flows, belonging to the same PHB traffic class might get
   other priority than other flows belonging to the same PHB traffic
   class. This difference in priority can be notified to the egress and
   ingress nodes either by the RESERVE message that carries the QSpec
   associated with the end-to-end QoS model, e.g.,, <Preemption
   Priority> & <Defending Priority> parameter, or by using a local
   defined policy. The priority value is kept in the reservation states,
   see Section 4.3, which might be used during admission control and/or
   severe congestion handling procedures. The terminated flows are
   selected from the flows having the same PHB traffic class as the PHB
   of the marked (as "encoded DSCP") and "affected DSCP" (when applied
   in the complete RMD domain) packets and (when the ingress/egress pair
   aggregated states are available) that are belonging to the same
   ingress/egress pair aggregate.

Bader, et al.                                                 [Page 56]


INTERNET-DRAFT                                                 RMD-QOSM

   For flows associated with the same PHB traffic class the priority of
   the flow plays a significant role. An example of calculating the
   number of flows associated with each priority class that have to be
   terminated is explained in Appendix A.1.2.

   For the flows (sessions) that have to be terminated, the QNE Egress
   node generates and sends an end-to-end NOTIFY message to the QNE
   Ingress node (its upstream stateful QoS-NSLP peer) to indicate the
   severe congestion in the communication path.


   The non-default values of the objects contained in the NOTIFY
   message MUST be set by the QNE Egress node as follows
   (see QoS-NSLP-RMF API described in [QoS-NSLP]):

   *  the values of the <INFO_SPEC> object is set by the standard
      QoS-NSLP protocol functions.

   * the INFO_SPEC object MUST include information that notifies that
     the end-to-end flow MUST be terminated. This information is as
     follows:

     Error Severity Class: Informational
     Error Code value: Congestion situation

   When the QNE edges maintain a per aggregate intra-domain QoS-NSLP
   operational state, see sections 4.3.1 then the QNE Edge has to
   calculate, per each aggregate intra-domain QoS-NSLP operational
   state, the total bandwidth that has to be terminated in order to
   solve the severe congestion. The total to be released bandwidth is
   calculated in the same way as in the situation that the QNE edges
   maintain per flow intra-domain QoS-NSLP operational states.
   Note that for the aggregated sessions that are affected, the QNE
   Egress node generates and sends one end-to-end NOTIFY message to the
   QNE Ingress node(its upstream stateful QoS-NSLP peer) to indicate the
   severe congestion in the communication path. Note that this end-to-
   end NOTIFY message is associated with one of the end-to-end sessions
   that is bound to the aggregated intra-domain QoS-NSLP operational
   state.

   The non-default values of the objects contained in the NOTIFY
   message MUST be set by the QNE Egress node in the same way as the
   ones used by the end-to-end NOTIFY message described above for the
   situation that the QNE Egress maintains a per flow intra-domain
   operational state. In addition to this the end-to-end NOTIFY MUST
   carry the RMD-Qspec, which contains a PDR container with a
   Parameter/Container ID = PDR_10, i.e., "PDR_Congestion_Report". The
   value of the <S> SHOULD be set. Furthermore, the value of the <PDR
   Bandwidth> parameter MUST contain the bandwidth, associated with the
   aggregated QoS-NSLP operational state, which has to be released.

 Bader, et al.                                                 [Page 57]


INTERNET-DRAFT                                                 RMD-QOSM

   Furthermore, the number of end-to-end sessions that have to be
   terminated will be calculated as in the situation that the QNE edges
   maintain per flow intra-domain QoS-NSLP operational states. Similarly
   for each, to be terminated, ongoing flow the egress will notify the
   ingress in the same way as in the situation that the QNE edges
   maintain per flow intra-domain QoS-NSLP operational states.

   Note that QNE egress SHOULD restore the original DSCP
   values of the remarked packets, otherwise multiple actions for the
   same event might occur. However, this value MAY be left in its
   remarking form if there is an SLA agreement between domains that a
   downstream domain handles the remarking problem.


   4.6.1.6.2.3 Operation in the Ingress nodes

   Upon receiving the (end-to-end) NOTIFY message, the QNE Ingress node
   resolves the severe congestion by a predefined policy, e.g., by
   refusing new incoming flows (sessions), terminating the affected and
   notified flows (sessions), and blocking their packets or shifting
   them to an alternative RMD traffic class (PHB). This operation is
   depicted in Figure 14, where the QNE Ingress, for each flow (session)
   to be terminated, receives a NOTIFY message that carries the
   "Congestion situation" error code.

   When the QNE Ingress node receives the end-to-end NOTIFY message, it
   associates this NOTIFY message with its bound intra-domain session,
   see Sections 4.3.2, 4.3.3. via the BOUND_SESSION_ID information
   included in the end-to-end per-flow QoS-NSLP state. The QNE Ingress
   uses the operation described in Section 4.6.1.5.2 to terminate the
   intra-domain session.

 QNE (Ingress)    QNE (Interior)        QNE (Interior)    QNE (Egress)

  user  |                  |                 |                  |
  data  |  user data       |                 |                  |
 ------>|----------------->|     user data   | user data        |
        |                  |---------------->S(# marked bytes)  |
        |                  |                 S----------------->|
        |                  |                 S(# unmarked bytes)|
        |                  |                 S----------------->|Term.
        |                 NOTIFY             S                  |flow?
        |<-----------------|-----------------S------------------|YES
        |RESERVE(RMD-QSpec:Tear=1,M=1,S=1)   S                  |
        | ---------------->|RESERVE(RMD-QSpec:T=1, M=1,S=1)     |
        |                  |                 S                  |
        |                  |---------------->S                  |
        |                  |       RESERVE(RMD-QSpec:Tear=1, M=1,S=1)
        |                  |                 S----------------->|

   Figure 14:  RMD severe congestion handling

 Bader, et al.                                                 [Page 58]


INTERNET-DRAFT                                                 RMD-QOSM

   Note that the above functionality applies to the RMD reservation-
   Based, see Section 4.3.3 and to both measurement-based admission
   control methods (i.e., congestion notification based on probing and
   the NSIS measurement-based admission control), see Section 4.3.2.

   In the case that the QNE edges support aggregated intra-domain QoS-
   NSLP operational states the following actions take place. The QNE
   Ingress MAY receive an end to end NOTIFY message with a PDR container
   that carries a <S> marked and a bandwidth value in the <PDR
   Bandwidth> parameter included in a "PDR_Congestion_Report" container.
   Furthermore the same end-to-end NOTIFY message carries an INFO_SPEC
   object with the "Congestion situation" error code.

   When the QNE Ingress node receives this end-to-end  NOTIFY message,
   it associates the NOTIFY message with the aggregated intra-domain
   QoS-NSLP operational state via the BOUND_SESSION_ID information
   included in the end-to-end per-flow QoS-NSLP operational state, see
   Section 4.3.1.

   The RMD-QOSM at the QNE Ingress node by using the
   total to be released bandwidth value included in the <PDR Bandwidth>
   parameter MUST reduce
   the bandwidth associated and reserved by the RMD aggregated session.
   This is accomplished by triggering the RMD modification for
   Aggregated reservations procedure described in Section 4.6.1.4.

   In addition to the above, the QNE Ingress MUST select a number of
   inter-domain (end-to-end) flows (sessions) that MUST be terminated.
   This is accomplished in the same way as in the situation that the QNE
   edges maintain per flow intra-domain QoS-NSLP operational states.

   The terminated end-to-end sessions are selected from the end-to-end
   sessions bound to the aggregated intra-domain QoS-NSLP operational
   state. Note that the end-to-end session associated with the received
   end-to-end NOTIFY message that notified the severe congestion MUST
   also be selected for termination.
   For the flows (sessions) that have to be terminated, the QNE Ingress
   node generates and sends an end-to-end NOTIFY message upstream
   towards the sender (QNI). The values carried by this message are:

   *  the values of the <INFO_SPEC> object is set by the standard
      QoS-NSLP protocol functions.

   * the INFO_SPEC object MUST include information that notifies that
     the end-to-end flow MUST be terminated. This information is as
     follows:

     Error Severity Class: Informational
     Error Code value: Congestion situation

Bader, et al.                                                 [Page 59]


INTERNET-DRAFT                                                 RMD-QOSM

4.6.1.7 Admission control using congestion notification based on probing

   The congestion notification function based on probing can be used to
   implement a simple measurement-based admission control within a
   Diffserv domain.  At interior nodes along the data path congestion
   notification thresholds are set in the measurement based admission
   control function for the traffic belonging to different PHBs. These
   interior nodes are not NSIS aware nodes.


4.6.1.7.1 Operation in Ingress nodes

   When an end-to-end reservation request (RESERVE) arrives at the
   Ingress node (QNE), see Figure 15, it is processed based on the
   procedures defined by the end-to-end QoS model.

   The DSCP field of the GIST datagram message that is used to transport
   this probe RESERVE message, SHOULD be marked with the same value of
   DSCP as the data path packets associated with the same session. In
   this way it is ensured that the end-to-end RESERVE (probe) packet
   passed through the node that it is congested. This feature is very
   useful when ECMP based routing is used to detect only flows that are
   passing through the congested router.

   When (end-to-end) RESPONSE message is received by the Ingress node,it
   will be processed based on the procedures defined by the end-to-end
   QoS model.


4.6.1.7.2 Operation in Interior nodes

   These Interior nodes are not needed to be NSIS aware nodes and they
   do not need to process NSIS functionality of NSIS messages. Note that
   the "not NSIS aware" nodes MUST be configured such that they can
   detect the congestion/severe congestion situations and remark packets
   in the same way as the "NSIS aware" nodes do.

   Using standard functionalities congestion notification thresholds are
   set for the traffic belonging to different PHBs, see Section 4.3.2.
   The end-to-end RESERVE message, see Figure 15, is used as a probe
   packet.

   The DSCP field of all data packets and of the GIST message carrying
   the RESERVE message will be re-marked when the corresponding
   "congestion notification" threshold is exceeded, see Section 4.3.2.
   Note that when the data rate is higher than the congestion
   notification threshold then also the data packets are remarked. An
   example of the detailed operation of this procedure is given in
   Appendix A.2.1.

Bader, et al.                                                 [Page 60]


INTERNET-DRAFT                                                 RMD-QOSM

4.6.1.7.3 Operation in Egress nodes

   As emphasised in Section 4.6.1.6.2.2, the egress node, by using the
   per flow end-to-end QoS-NSLP states, can derive which flows are using
   the same PHB and are sent by the same ingress.

   For each ingress, the egress SHOULD generate an ingress/egress pair
   aggregated (RMF) reservation state for each supported PHB. Note that
   this aggregated reservation state does not require that also an
   aggregated intra-domain QoS-NSLP operational state is needed.

   In Appendix A.2.2 an example is described how and when a (probe)
   RESERVE message that arrives at the egress, is admitted or rejected.

   If the request is rejected then the Egress node SHOULD
   generate an (end-to-end) RESPONSE message to notify that the
   reservation is unsuccesfull. In particular it will generate an
   INFO_SPEC object of:
     Error Severity Class: Transient Failure
     Error Code value: Reservation failure

   The QSpec that was carried by the end to end RESERVE belonging to
   the same session as this end to end RESPONSE is included in this
   message. The parameters included in the QSpec <QoS Reserved> object
   are copied from the original <QoS Desired> values. The "E" flag
   associated with the <QoS Reserved> object and the "E" flag associated
   with <Bandwidth> parameter are also set. This RESPONSE message will
   be sent to the Ingress node and it will be processed based on the
   end-to-end QoS model.

   Note that QNE egress SHOULD restore the original DSCP values of the
   remarked packets, otherwise multiple actions for the same event might
   occur. However, this value MAY be left in its remarking form if there
   is an SLA agreement between domains that a downstream domain handles
   the remarking problem. Note that the break "B" flag carried by the
   end-to-end RESERVE message MUST NOT be set.

Bader, et al.                                                 [Page 61]


INTERNET-DRAFT                                                 RMD-QOSM

QNE (Ingress)          Interior          Interior       QNE (Egress)
                    (not NSIS aware) (not NSIS aware)
  user  |                  |                 |                  |
  data  |  user data       |                 |                  |
 ------>|----------------->|     user data   |                  |
        |                  |---------------->| user data        |
        |                  |                 |----------------->|
  user  |                  |                 |                  |
  data  |  user data       |                 |                  |
 ------>|----------------->|     user data   | user data        |
        |                  |---------------->S(# marked bytes)  |
        |                  |                 S----------------->|
        |                  |                 S(# unmarked bytes)|
        |                  |                 S----------------->|
        |                  |                 S                  |
RESERVE |                  |                 S                  |
------->|                  |                 S                  |
        |----------------------------------->S                  |
        |                  |           RESERVE(re-marked DSCP in GIST)
        |                  |                 S----------------->|
        |                  |RESPONSE(unsuccessful INFO-SPEC)    |
        |<------------------------------------------------------|
 RESPONSE(unsuccessful INFO-SPEC)            |                  |
 <------|                  |                 |                  |

   Figure 15:  Using RMD congestion notification function for admission
               control based on probing

4.6.2  Bi-directional operation

   RMD-QOSM assumes that asymmetric routing MAY be applied in the RMD
   domain.  Combined sender-receiver initiated reservation cannot be
   efficiently done in the RMD domain because upstream NTLP states are
   not stored in Interior routers.  Therefore, the bi-directional
   operation SHOULD be performed by two sender-initiated reservations
   (sender&sender).  We assume that the QNE edge nodes are common for
   both upstream and downstream directions, therefore, the two
   reservations/sessions can be bound at the QNE edge nodes. Note that
   if this is not the case then the bi-directional procedure could be
   managed and maintained by nodes located outside the RMD domain, by
   using other procedures than the ones defined in RMD-QOSM.

   This (intra-domain) bi-directional sender&sender procedure can then
   be applied between the QNE edges (QNE Ingress and QNE Egress) nodes
   of the RMD QoS signaling model.  In the situation a security
   association exists between the QNE Ingress and QNE Egress nodes (see
   Figure 15), and the QNE Ingress node has the REQUIRED <Bandwidth>
   parameters for both directions, i.e., QNE Ingress towards QNE Egress
   and QNE Egress towards QNE Ingress, then the QNE Ingress MAY include
   both <Bandwidth> parameters (needed for both directions) into the
   RMD-QSpec within a RESERVE message.  In this way the QNE Egress node
   is able to use the QoS parameters needed for the "Egress towards
   Ingress" direction (QoS-2).  The QNE Egress is then able to create a

Bader, et al.                                                 [Page 62]


INTERNET-DRAFT                                                 RMD-QOSM

   RESERVE with the right QoS parameters included in the QSpec, i.e.,
   RESERVE (QoS-2). Both directions of the flows are bound by inserting
   <BOUND_SESSION_ID> objects at the QNE Ingress and QNE Egress, which
   will be carried by bound end-to-end RESERVE messages.

     |------ RESERVE (QoS-1, QoS-2)----|
     |                                 V
     |           Interior/stateless QNEs
                 +---+     +---+
        |------->|QNE|-----|QNE|------
        |        +---+     +---+     |
        |                            V
      +---+                        +---+
      |QNE|                        |QNE|
      +---+                        +---+
         ^                           |
      |  |       +---+     +---+     V
      |  |-------|QNE|-----|QNE|-----|
      |          +---+     +---+
   Ingress/                         Egress/
   statefull QNE                    statefull QNE
                                     |
   <--------- RESERVE (QoS-2) -------|

   Figure 16: The intra-domain bi-directional reservation scenario in
   the RMD domain


   Note that it is RECOMMENDED that the QNE implementations of RMD-QOSM
   process the QoS-NSLP signaling messages with a higher priority than
   data packets. This can be accomplished as described in Section 3.3.4
   in [QoS-NSLP] and the QoS-NSLP-RMF API [QoS-NSLP].

   A bidirectional reservation, within the RMD domain, is indicated by
   the PHR <B> and PDR <B> flags, which are set in all messages. In this
   case two BOUND_SESSION_ID objects SHOULD be used.

   When the QNE edges maintain per-flow intra-domain QoS-NSLP
   operational states then the end-to-end RESERVE message carries two
   BOUND_SESSION_IDs. One BOUND_SESSION_ID carries the SESSION_ID of the
   tunneled intra-domain (per-flow) session that is using a BINDING_CODE
   with value set to code (Tunneled and end-to-end sessions).  Another
   BOUND_SESSION_ID carries the SESSION_ID of the bound bidirectional
   end-to-end session. The BINDING_CODE associated with this
   BOUND_SESSION_ID is set to code (Bi-directional sessions).

   When the QNE edges maintain aggregated intra-domain QoS-NSLP
   operational states then the end-to-end RESERVE message carries two
   BOUND_SESSION_IDs. One BOUND_SESSION_ID carries the SESSION_ID of the
   tunneled aggregated intra-domain session that is using a BINDING_CODE
   with value set to code (Aggregated sessions).  Another
   BOUND_SESSION_ID carries the SESSION_ID of the bound bidirectional
   end-to-end session. The BINDING_CODE associated with this
   BOUND_SESSION_ID is set to code (Bi-directional sessions).

Bader, et al.                                                 [Page 63]


INTERNET-DRAFT                                                 RMD-QOSM

   The intra-domain and end-to-end QoS-NSLP operational states are
   initiated/modified depending on the binding type, see Section 4.3.1,
   4.3.2, 4.3.3.

   If no security association exists between the QNE Ingress and QNE
   Egress nodes the bi-directional reservation for the sender&sender
   scenario in the RMD domain SHOULD use the scenario specified in
   [QoS-NSLP] as "Bi-directional reservation for sender&sender
   scenario". This is because in this scenario the RESERVE message sent
   from QNE Ingress to QNE Egress does not have to carry the QoS
   parameters needed for the "Egress towards Ingress" direction (QoS-2).

   In the following sections it is considered that the QNE
   edge nodes are common for both upstream and downstream directions
   and therefore, the two reservations/sessions can be bound at the
   QNE edge nodes.  Furthermore, it is considered that a security
   association exists between the QNE Ingress and QNE Egress nodes,
   and the QNE Ingress node has the REQUIRED <Bandwidth> parameters
   for both directions, i.e., QNE Ingress towards QNE Egress and
   QNE Egress towards QNE Ingress.


4.6.2.1 Successful and unsuccessful reservations

   This section describes the operation of the RMD-QOSM where a RMD
   Intra-domain bi-directional reservation operation, see Figure 16 and
   Section 4.6.2, is either successfully or unsuccessfully accomplished.

   The bi-directional successful reservation is similar to a
   combination of two unidirectional successful reservations that are
   accomplished in opposite directions, see Figure 17. The main
   differences of the bi-directional successful reservation procedure
   with the combination of two unidirectional successful reservations
   accomplished in opposite directions are as follows. Note also that
   the intra-domain and end-to-end QoS-NSLP operational states generated
   and maintained by the end-to-end RESERVE messages contain, compared
   to the unidirectional reservation scenario, a different
   BOUND_SESSION_ID data structure, see Section 4.3.1, 4.3.2, 4.3.3.
   In this scenario the intra-domain RESERVE message sent by the QNE
   Ingress node towards the QNE Egress node, is denoted in Figure 17 as
   RESERVE (RMD-QSpec): "forward".  The main differences between the
   intra-domain RESERVE (RMD-QSpec):"forward" message used for the bi-
   directional successful reservation procedure and a RESERVE (RMD-
   QSpec) message used for the unidirectional successful reservation are
   as follows (see QoS-NSLP-RMF API described in [QoS-NSLP]):

   *  the RII object MUST NOT be included in the message. This is
      because no RESPONSE message is REQUIRED.

   *  the <B> bit of the PHR container indicates a bi-directional
      reservation and it MUST be set to "1".

Bader, et al.                                                 [Page 64]


INTERNET-DRAFT                                                 RMD-QOSM

   *  the PDR container is also included into the RESERVE(RMD-QSpec):
      "forward" message. The value of the Parameter/Container ID is
      "PDR_4", i.e., "PDR_Reservation_Request".  Note that the response
      PDR container sent by a QNE Egress to a QNE Ingress node is not
      carried by an end-to-end RESPONSE message, but it is carried by an
      intra-domain RESERVE message that is sent by the QNE Egress node
      towards the QNE Ingress node (denoted in Figure 16 as
      RESERVE(RMD-QSpec):"reverse").

   *  the <B> PDR bit indicates a bi-directional reservation and is set
      to "1".

   *  the <PDR Bandwidth> field specifies the
      requested bandwidth that has to be used by the QNE Egress node to
      initiate another intra-domain RESERVE message in the reverse
      direction.

   The RESERVE(RMD-QSpec):"reverse" message is initiated by the QNE
   Egress node at the moment that the RESERVE(RMD-QSpec):"forward"
   message is successfully processed by the QNE Egress node.
   The main differences between the RESERVE(RMD-QSpec):"reverse"
   message used for the bi-directional successful reservation procedure
   and a RESERVE(RMD-QSpec) message used for the unidirectional
   successful reservation are as follows:

QNE (Ingress)   QNE (int.)    QNE (int.)    QNE (int.)   QNE (Egress)
NTLP stateful  NTLP st.less  NTLP st.less  NTLP st.less  NTLP stateful
    |                |               |               |              |
    |                |               |               |              |
    |RESERVE(RMD-QSpec)              |               |              |
    |"forward"       |               |               |              |
    |                |    RESERVE(RMD-QSpec):        |              |
    |--------------->|    "forward"  |               |              |
    |                |------------------------------>|              |
    |                |               |               |------------->|
    |                |               |               |              |
    |                |               |RESERVE(RMD-QSpec)            |
    |      RESERVE(RMD-QSpec)        | "reverse"     |<-------------|
    |      "reverse" |               |<--------------|              |
    |<-------------------------------|               |              |

      Figure 17: Intra-domain signaling operation for successful
                 bi-directional reservation

   *  the RII object is not included in the message. This is because no
      RESPONSE message is REQUIRED;

   *  the value of the <Bandwidth> parameter is set equal to the value
      of the <PDR Bandwidth> field included in the
      RESERVE(RMD-QSpec):"forward" message that triggered the
      generation of this RESERVE(RMD-QSpec): "reverse" message;

   *  the <B> bit of the PHR container indicates a bi-directional
      reservation and is set to "1";

Bader, et al.                                                 [Page 65]


INTERNET-DRAFT                                                 RMD-QOSM

   *  the PDR container is included into the
      RESERVE(RMD-QSpec):"reverse" message.  The value of the
      Parameter/Container ID is "PDR_7", i.e., "PDR_Reservation_Report";

   *  the <B> PDR bit indicates a bi-directional reservation and is
      set to "1".

   Figure 18 and Figure 19 show the flow diagrams used in case of a
   unsuccessful bi-directional reservation.  In Figure 18 it
   is considered that the QNE that is not able to support the
   requested <Bandwidth> is located in the direction QNE Ingress
   towards QNE Egress.  In Figure 19 it is considered that the
   QNE that is not able to support the requested <Bandwidth> is
   located in the direction QNE Egress towards QNE Ingress.

   The main differences between the bi-directional unsuccessful
   procedure shown in Figure 18 and the bi-directional successful
   procedure are as follows:

   *  the QNE node that is not able to reserve resources for a
      certain request is located in the "forward" path, i.e., path
      from QNE Ingress towards the QNE Egress.

   *  the QNE node that is not able to support the requested
      <Bandwidth> it MUST mark the <M> bit, i.e., set to value "1", of
      the RESERVE(RMD-QSpec): "forward".

QNE(Ingress)   QNE (int.)    QNE (int.)    QNE (int.)    QNE (Egress)
NTLP stateful  NTLP st.less  NTLP st.less  NTLP st.less  NTLP stateful
    |                |             |              |               |
    |RESERVE(RMD-QSpec):           |              |               |
    |  "forward"     |  RESERVE(RMD-QSpec):       |               |
    |--------------->|  "forward"  |              M RESERVE(RMD-QSpec):
    |                |--------------------------->M  "forward-M marked"
    |                |             |              M-------------->|
    |                |           RESPONSE(PDR)    M               |
    |                |        "forward - M marked"M               |
    |<------------------------------------------------------------|
    |RESERVE(RMD-QSpec, K=0)       |              M               |
    |"forward - T tear"            |              M               |
    |--------------->|             |              M               |
    |                    RESERVE(RMD-QSpec, K=1)  M               |
    |                |   "forward - T tear"       M               |
    |                |--------------------------->M               |
    |                |                  RESERVE(RMD-QSpec, K=1)   |
    |                |                 "forward - T tear"         |
    |                |                            M-------------->|

   Figure 18: Intra-domain signaling operation for unsuccessful
              bi-directional reservation (rejection on path QNE(Ingress)
              towards QNE(Egress))

Bader, et al.                                                 [Page 66]


INTERNET-DRAFT                                                 RMD-QOSM

   The operation for this type of unsuccessful bi-directional
   reservation is similar to the operation for unsuccessful uni-
   directional reservation shown in Figure 9.

   The main differences between the bi-directional unsuccessful
   procedure shown in Figure 19 and the in bi-directional successful
   procedure are as follows:

   *  the QNE node that is not able to reserve resources for a
      certain request is located in the "reverse" path, i.e., path
      from QNE Egress towards the QNE Ingress.

   *  the QNE node that is not able to support the requested
      <Bandwidth> it MUST mark the <M> bit, i.e., set to value "1",
      the RESERVE(RMD-QSpec):"reverse".

QNE (Ingress)    QNE (int.)    QNE (int.)    QNE (int.)    QNE (Egress)
NTLP stateful   NTLP st.less  NTLP st.less  NTLP st.less   NTLP stateful
    |                |                |                |              |
    |RESERVE(RMD-QSpec)               |                |              |
    |"forward"       |  RESERVE(RMD-QSpec):            |              |
    |--------------->|  "forward"     |           RESERVE(RMD-QSpec): |
    |                |-------------------------------->|"forward"     |
    |                |   RESERVE(RMD-QSpec):           |------------->|
    |                |    "reverse"   |                |              |
    |                |              RESERVE(RMD-QSpec) |              |
    |    RESERVE(RMD-QSpec):          M      "reverse" |<-------------|
    |   "reverse - M marked"          M<---------------|              |
    |<--------------------------------M                |              |
    |                |                M                |              |
    |RESERVE(RMD-QSpec, K=0):         M                |              |
    |"forward - T tear"               M                |              |
    |--------------->|  RESERVE(RMD-QSpec, K=0):       |              |
    |                |  "forward - T tear"             |              |
    |                |-------------------------------->|              |
    |                |                M                |------------->|
    |                |                M         RESERVE(RMD-QSpec, K=0):
    |                |                M             reverse - T tear" |
    |                |                M                |<-------------|
    |                                 M RESERVE(RMD-QSpec, K=1)       |
    |                |                M "forward - T tear"            |
    |                |                M<---------------|              |
    |          RESERVE(RMD-QSpec, K=1)M                |              |
    |          "forward - T tear"     M                |              |
    |<--------------------------------M                |              |

   Figure 19: Intra-domain signaling normal operation for unsuccessful
             bi-directional reservation (rejection on path QNE(Egress)
             towards QNE(Ingress)

Bader, et al.                                                 [Page 67]


INTERNET-DRAFT                                                 RMD-QOSM

   *  the QNE Ingress uses the information contained in the received
      PHR and PDR containers of the RESERVE(RMD-QSpec): "reverse" and
      generates a tear intra-domain RESERVE(RMD-QSpec):
      "forward - T tear" message.  This message carries a
      "PHR_Release_Request" and a "PDR_Release_Request" control
      information.  This message is sent to QNE Egress node.
      The QNE Egress node uses the information contained in the
      "PHR_Release_Request" and the "PDR_Release_Request" control
      info containers to generate a RESERVE(RMD-QSpec):"reverse - T
      tear" message that is sent towards the QNE Ingress node.


4.6.2.2 Refresh reservations

   This section describes the operation of the RMD-QOSM where a RMD
   intra-domain bi-directional refresh reservation operation is
   accomplished.
   The refresh procedure in case of RMD reservation-based method
   follows a similar scheme as the successful reservation procedure,
   described in Section 4.6.2.1, and depicted in Figure 17 and the
   way of how the refresh process of the reserved resources is
   maintained, is similar to the refresh process used for the intra-
   domain uni-directional reservations (see Section 4.6.1.3).

   Note that the RMD traffic class refresh periods used by the bound bi-
   directional sessions MUST be equal in all QNE edge and QNE Interior
   nodes.

   The main differences between the RESERVE(RMD-QSpec):"forward"
   message used for the bi-directional refresh procedure
   and a RESERVE(RMD-QSpec):"forward" message used for the bi-
   directional successful reservation procedure are as follows:

   *  the value of the Parameter/Container ID of the PHR container is
      "PHR_2", i.e., "PHR_Refresh_Update".

   *  the value of the Parameter/Container ID of the PDR container is
      "PDR_5", i.e., "PDR_Refresh_Request".

   The main differences between the RESERVE(RMD-QSpec):"reverse"
   message used for the bi-directional refresh procedure and the RESERVE
   (RMD-QSpec): "reverse" message used for the bi-directional successful
   reservation procedure are as follows:

   *  the value of the Parameter/Container ID of the PHR container is
      "PHR_2", i.e., "PHR_Refresh_Update".

   *  the value of the Parameter/Container ID of the PDR container is
      "PDR_8", i.e., "PDR_Refresh_Report".


Bader, et al.                                                 [Page 68]


INTERNET-DRAFT                                                 RMD-QOSM

4.6.2.3 Modification of aggregated intra-domain QoS-NSLP operational
reservation states

   This section describes the operation of the RMD-QOSM where RMD
   intra-domain bi-directional QoS-NSLP aggregated reservation states
   have to be modified.

   In the case when the QNE edges maintain, for the RMD QoS model,
   QoS-NSLP aggregated reservation states and if such an aggregated
   reservation has to be modified (see Section 4.3.1) then similar
   procedures to Section 4.6.1.4. are applied. In particular:

   * When the modification request requires an increase of the reserved
   resources, the QNE Ingress node MUST include the corresponding value
   into the <Bandwidth> parameter of the "RMD-QOSM object combination"
   (i.e., the RMD-QOSM <QoS Desired>), which is sent together with a
   "PHR_Resource_Request" control information.  If a QNE edge or QNE
   Interior node is not able to reserve the number of requested
   resources, then the "PHR_Resource_Request" associated with the
   <Bandwidth> parameter MUST be marked.  In this situation the RMD
   specific operation for unsuccessful reservation will be applied (see

   Section 4.6.2.1). Note that the value of the <PDR Bandwidth>
   parameter, which is sent within a "PDR_Reservation_Request"
   container, represents the increase of the reserved resources in the
   "reverse" direction.

   * When the modification request requires a decrease of the
   reserved resources, the QNE Ingress node MUST include this value
   into the <Bandwidth> parameter of the "RMD-QOSM object combination"
   (i.e., the RMD-QOSM <QoS Desired>). Subsequently an RMD release
   procedure SHOULD be accomplished (see Section 4.6.2.4). Note that the
   value of the <PDR Bandwidth> parameter, which is sent within a
   "PDR_Release_Request" container, represents the decrease of the
   reserved resources in the "reverse" direction.


4.6.2.4 Release procedure

   This section describes the operation of the RMD-QOSM where a RMD
   intra-domain bi-directional reservation release operation is
   accomplished. The message sequence diagram used in this procedure is
   similar to the one used by the successful reservation procedures,
   described in Section 4.6.2.1, and depicted in Figure 17. However, the
   way of how the release of the reservation is accomplished, is similar
   to the RMD release procedure used for the intra-domain uni-
   directional reservations (see Section 4.6.1.5 and Figure 18 and
   Figure 19).

   The main differences between the RESERVE (RMD-QSpec):
   "forward" message used for the bi-directional release procedure
   and a RESERVE (RMD-QSpec): "forward" message used for the bi-
   directional successful reservation procedure are as follows:

Bader, et al.                                                 [Page 69]


INTERNET-DRAFT                                                 RMD-QOSM

   *  the value of the Parameter/Container ID of the PHR container is
      "PHR_3", i.e."PHR_Release_Request";

   *  the value of the Parameter/Container ID of the PDR container is
      "PDR_6", i.e., "PDR_Release_Request";

   The main differences between the RESERVE (RMD-QSpec): "reverse"
   message used for the bi-directional release procedure and the RESERVE
   (RMD-QSpec): "reverse" message used for the bi-directional successful
   reservation procedure are as follows:

   *  the value of the Parameter/Container ID of the PHR container is
      "PHR_3", i.e., "PHR_Release_Request";

   *  the PDR container is not included in the RESERVE (RMD-QSpec):
      "reverse" message.


4.6.2.5 Severe congestion handling

   This section describes the severe congestion handling operation used
   in combination with RMD intra-domain bi-directional reservation
   procedures. This severe congestion handling operation is similar to
   the one described in Section 4.6.1.6.

4.6.2.5.1 Severe congestion handling by the RMD-QOSM bi-directional
          refresh procedure

   This procedure is similar to the severe congestion handling procedure
   described in Section 4.6.1.6.1. The difference is related to how the
   refresh procedure is accomplished, see Section 4.6.2.2 and to how the
   flows are terminated, see Section 4.6.2.4.

4.6.2.5.2 Severe congestion handling by proportional data packet marking

   This section describes the severe congestion handling by proportional
   data packet marking when this is combined with a RMD intra-domain
   bi-directional reservation procedure. Note that the detection and
   marking/remarking functionality described in this section and used by
   Interior nodes, applies to NSIS aware, but also to NSIS unaware
   nodes. This means however, that the "not NSIS aware" Interior nodes
   MUST be configured such that they can detect the congestion
   situations and remark packets in the same way as the Interior "NSIS
   aware" nodes do.

Bader, et al.                                                 [Page 70]


INTERNET-DRAFT                                                 RMD-QOSM

QNE(Ingress)   QNE (int.)    QNE (int.)    QNE (int.)    QNE (Egress)
NTLP stateful  NTLP st.less  NTLP st.less  NTLP st.less  NTLP stateful
user|                |             |              |               |
data|    user        |             |              |               |
--->|    data        | user data   |              |user data      |
    |--------------->|             |              S               |
    |                |--------------------------->S (#marked bytes)
    |                |             |              S-------------->|
    |                |             |              S(#unmarked bytes)
    |                |             |              S-------------->|Term
    |                |             |              S               |flow?
    |                |          NOTIFY (PDR)      S               |YES
    |<------------------------------------------------------------|
    |RESERVE(RMD-QSpec)            |              S               |
    |"forward - T tear"            |              S               |
    |--------------->|             |           RESERVE(RMD-QSpec):|
    |                |--------------------------->S"forward - T tear"
    |                |             |              S-------------->|
    |                |             |          RESERVE(RMD-QSpec): |
    |                |             |           "reverse - T tear" |
    | RESERVE(RMD-QSpec):          |              |<--------------|
    |"reverse - T tear"            |<-------------S               |
    |<-----------------------------|              S               |

Figure 20: Intra-domain RMD severe congestion handling for
           bi-directional reservation (congestion on path QNE(Ingress)
           towards QNE(Egress))

   This procedure is similar to the severe congestion handling procedure
   described in Section 4.6.1.6.2. The main difference is related to the
   location of the severe congested node, i.e. "forward" or "reverse"
   path. Note that when a severe congestion situation occurs on
   e.g. on a forward path, and flows are terminated to solve the severe
   congestion in forward path, then the reserved bandwidth associated
   with the terminated bidirectional flows will also be released.
   Therefore, a careful selection of the flows that have to be
   terminated SHOULD take place. An example of such a selection is
   given in Appendix A.3.1.

   Furthermore, a special case of this operation is associated to the
   severe congestion situation occurring simultaneously on the forward
   and reverse paths. An example of this operation is given in Appendix
   A.3.2.

    Simulation results associated with these procedures can be found in
   [DiKa08].

Bader, et al.                                                 [Page 71]


INTERNET-DRAFT                                                 RMD-QOSM

QNE (Ingress)    QNE (int.)    QNE (int.)    QNE (int.)    QNE (Egress)
NTLP stateful   NTLP st.less  NTLP st.less  NTLP st.less   NTLP stateful
user|                |                |           |               |
data|    user        |                |           |               |
--->|    data        | user data      |           |user data      |
    |--------------->|                |           |               |
    |                |--------------------------->|user data      |user
    |                |                |           |-------------->|data
    |                |                |           |               |--->
    |                |                |  user     |               |<---
    |   user data    |                |  data     |<--------------|
    | (#marked bytes)|                S<----------|               |
    |<--------------------------------S           |               |
    | (#unmarked bytes)               S           |               |
Term|<--------------------------------S           |               |
Flow?                |                S           |               |
YES |RESERVE(RMD-QSpec):              S           |               |
    |"forward - T tear"               s           |               |
    |--------------->|  RESERVE(RMD-QSpec):       |               |
    |                |  "forward - T tear"        |               |
    |                |--------------------------->|               |
    |                |                S           |-------------->|
    |                |                S         RESERVE(RMD-QSpec):
    |                |                S       "reverse - T tear"  |
    |      RESERVE(RMD-QSpec)         S           |<--------------|
    |      "reverse - T tear"         S<----------|               |
    |<--------------------------------S           |               |

   Figure 21: Intra-domain RMD severe congestion handling for
           bi-directional reservation (congestion on path QNE(Egress)
           towards QNE(Ingress))

   Figure 20 shows the scenario where the severe congested node is
   located in the "forward" path. The QNE Egress node has to generate an
   end-to-end NOTIFY(PDR) message. In this way the QNE Ingress will be
   able to receive the (#marked and #unmarked) that were measured by the
   QNE Egress node on the congested "forward path". Note that in this
   situation it is assumed that the "reverse path" is not congested.
   This scenario is very similar to the
   severe congestion handling scenario described in Section 4.6.1.6.2
   and shown in Figure 14. The difference is related to the release
   procedure, which is accomplished in the same way as described in
   Section 4.6.2.4.

Bader, et al.                                                 [Page 72]


INTERNET-DRAFT                                                 RMD-QOSM

   Figure 21 shows the scenario where the severe congested node is
   located in the "reverse" path. Note that in this situation it is
   assumed that the "forward path" is not congested. The main difference
   between this scenario and the scenario shown in Figure 20 is that no
   end-to-end NOTIFY(PDR) message has to be generated by the QNE Egress
   node. This is because now the severe congestion occurs on the
   "reverse path" and the QNE Ingress node receives the (#marked and
   #unmarked) user data passing through the severely congested "reverse
   path". The QNE Ingress node will be able to calculate the number of
   flows that have to be terminated or forwarded in a lower priority
   queue.

   For the flows that have to be terminated a release procedure, see
   Section 4.6.2.4, is initiated to release the reserved resources
   on the "forward" and "reverse" paths.


4.6.2.6 Admission control using congestion notification based on
          probing

   This section describes the admission control scheme that uses the
   congestion notification function based on probing when RMD
   intra-domain bi-directional reservations are supported.

QNE(Ingress)    Interior    QNE (int.)      Interior      QNE (Egress)
NTLP stateful not NSIS aware not NSIS aware not NSIS aware NTLP stateful
user|                |             |              |               |
data|                |             |              |               |
--->|                | user data   |              |user data      |
    |-------------------------------------------->S (#marked bytes)
    |                |             |              S-------------->|
    |                |             |              S(#unmarked bytes)
    |                |             |              S-------------->|
    |                |             |              S               |
    |                |           RESERVE(re-marked DSCP in GIST)):|
    |                |             |              S               |
    |-------------------------------------------->S               |
    |                |             |              S-------------->|
    |                |             |              S               |
    |                |          RESPONSE(unsuccessful INFO-SPEC)  |
    |<------------------------------------------------------------|
    |                |             |              S               |


   Figure 22: Intra-domain RMD congestion notification based on probing
              for bi-directional admission control (congestion on path
              from QNE(Ingress) towards QNE(Egress))

   This procedure is similar to the congestion notification for
   admission control procedure described in Section 4.6.1.7. The main
   difference is related to the location of the severe congested node,
   i.e., "forward" path (i.e., path between QNE Ingress towards QNE
   Egress) or "reverse" path (i.e., path between QNE Egress towards
   QNE Ingress).

Bader, et al.                                                 [Page 73]


INTERNET-DRAFT                                                 RMD-QOSM

   Figure 22 shows the scenario where the severe congested node is
   located in the "forward" path. The functionality of providing
   admission control is the same as the one described in Section
   4.6.1.7, Figure 15.

   Figure 23 shows the scenario where the congested node is located in
   the "reverse" path. The probe RESERVE message sent in the "forward"
   direction will not be affected by the severe congested node, while
   the DSCP value in the IP header of any packet of the "reverse"
   direction flow and also of the GIST message that carries the
   probe RESERVE message sent in the "reverse" direction will be
   remarked by the congested node. The QNE ingress is in this way
   notified that a congestion occurred in the network and therefore it
   is able to refuse the new initiation of the reservation.

QNE (Ingress)    Interior    QNE (int.)     Interior       QNE (Egress)
NTLP stateful not NSIS aware  NTLP st.less not NSIS aware NTLP stateful
user|                |                |           |               |
data|                |                |           |               |
--->|                | user data      |           |               |
    |-------------------------------------------->|user data      |user
    |                |                |           |-------------->|data
    |                |                |           |               |--->
    |                |                |           |               |user
    |                |                |           |               |data
    |                |                |           |               |<---
    |                S                | user data |               |
    |                S  user data     |<--------------------------|
    |   user data    S<---------------|           |               |
    |<---------------S                |           |               |
    |  user data     S                |           |               |
    | (#marked bytes)S                |           |               |
    |<---------------S                |           |               |
    |                S           RESERVE(unmarked DSCP in GIST)): |
    |                S                |           |               |
    |----------------S------------------------------------------->|
    |                S          RESERVE(re-marked DSCP in GIST)   |
    |                S<-------------------------------------------|
    |<---------------S                |           |               |

   Figure 23: Intra-domain RMD congestion notification for
           bi-directional admission control (congestion on path
           QNE(Egress) towards QNE(Ingress))

   Note that the "not NSIS aware" Interior nodes MUST be configured
   such that they can detect the congestion/severe congestion
   situations and remark packets in the same way as the Interior
   "NSIS aware" nodes do.

4.7 Handling of additional errors

   During the QSpec processing, additional errors MAY occur. The way
   in which these additional errors are handled and notified is
   specified in [QSP-T] and [QoS-NSLP].

Bader, et al.                                                 [Page 74]


INTERNET-DRAFT                                                 RMD-QOSM

5.  Security Considerations

   A router implementing a QoS signaling protocol can, similar to a
   router without QoS signaling, do a lot of harm to a system. If taken
   over by an adversary, a router can delay, drop, inject, duplicate or
   modify packets. Additional threats are, however, introduced with new
   protocols and they are subject for a discussion below.

   The RMD-QOSM aims to be very lightweight signaling with regard to
   the number of signaling message roundtrips and the amount of state
   established at involved signaling nodes with and without reduced
   state on QNEs. This implies the usage of the Datagram Mode which
   does not allow channel security to be used. As such, RMD signaling is
   targeted towards intra-domain signaling only. Therefore, the security

   considerations described in this section apply to all the supported
   RMD-QOSM/QoS-NSLP signalling schemes, see Section 3.2.3.

       QNE             QNE             QNE            QNE
     Ingress         Interior        Interior        Egress
 NTLP stateful  NTLP stateless  NTLP stateless  NTLP stateful
        |               |               |              |
        | RESERVE (1)   |               |              |
        +--------------------------------------------->|
        | RESERVE` (2)  |               |              |
        +-------------->|               |              |
        |               | RESERVE`      |              |
        |               +-------------->|              |
        |               |               | RESERVE`     |
        |               |               +------------->|
        |               |               | RESPONSE` (2)|
        |<---------------------------------------------+
        |               |               | RESPONSE (1) |
        |<---------------------------------------------+

                 Figure 24: RMD message exchange

   In the context of RMD-QOSM signaling a classification between
   on-path adversaries and off-path adversaries needs to be made.
   Furthermore, it might be necessary to differentiate between off-path
   nodes that never participate in the RMD signaling exchange and nodes

   that are only off-path with regard to a specific signaling session
   whereby routing asymmetry might even mean that the downstream and the
   upstream signaling direction matters for this classification.

   Note that RMD always uses the message exchange shown in Figure 24
   even if there is no end-to-end signaling session. If the RMD-QOSM is
   triggered based on an E2E signaling exchange then the RESERVE message
   is created by a node outside the RMD domain and will subsequently
   travel further on (e.g., to the data receiver). Such an exchange is
   shown in Figure 3. As such, an evaluation of RMD`s security MUST
   always been seen as a combination of the two signaling sessions, (1)
   and (2) of Figure 24.

Bader, et al.                                                 [Page 75]


INTERNET-DRAFT                                                 RMD-QOSM

   The following security requirements are set as goals for the
   intra-domain communication, namely:

*  Nodes, which are never supposed to participate in the NSIS signaling
   exchange, MUST NOT interfere with QNE Interior nodes. Off-path
   nodes (off-path with regard to the path taken by a particular
   signaling message exchange) MUST NOT be able to interfere with
   other on-path signaling nodes.

*  The actions allowed by a QNE Interior node SHOULD be minimal (i.e.,
   only those specified by the RMD-QOSM). For example, only the QNE
   Ingress and the QNE Egress nodes are allowed to initiate certain
   signaling messages. QNE Interior nodes are, for example, allowed to
   modify certain signaling message payloads.

   Note that the term `interfere` refers to all sorts of security
   threats, such as denial of service, spoofing, replay, signaling
   message injection, etc.

   If we assume that the RESERVE/RESPONSE is sent with hop-by-hop
   channel security provided by GIST and protected between the QNE
   Ingress and the QNE Egress node then we can
   be sure that the payloads of these messages MUST be authenticated,
   integrity, replay protected and encrypted. Encryption is necessary to
   prevent an adversary that is located along the path of the RESERVE
   message to learn information about the session that can later be used
   to inject a valid RESERVE`. The following messages need to relate to
   each other to make sure that the occurrence of one message is not
   without the other one:

   a) the RESERVE and the RESERVE` relate to each other at the QNE
      Egress and

   b) the RESPONSE and the RESERVE relate to each other at the QNE
      Ingress and

   c) the RESERVE` and the RESPONSE` relate to each other. The RII is
   carried in the RESERVE` message and the RESPONSE` message that is
   generated by the QNE Egress node contains the same RII as the
   RESERVE`. The RII can be used by the QNE Ingress to match the
   RESERVE` with the RESPONSE`. The QNE Egress is able to determine
   whether the RESERVE` was created by the QNE Ingress node since the
   intra-domain session, which sent the RESERVE`, is bound to an end-to-
   end session via the BOUND_SESSION_ID value included in the intra-
   domain QoS-NSLP operational state maintained at the QNE Egress.

Bader, et al.                                                 [Page 76]


INTERNET-DRAFT                                                 RMD-QOSM

   The RESERVE and the RESERVE` message are tied together using the
   BOUND_SESSION_ID(s) maintained by the intra-domain and end-to-end
   QoS-NSLP operational states maintained at the QNE edges, see Section
   4.3.1, 4.3.2, 4.3.3. Hence, there cannot be a RESERVE` without a
   corresponding RESERVE. The SESSION_ID can fulfill this purpose quite
   well if the aim is to provide protection against off-path adversaries
   that do not see the SESSION_ID carried in the RESERVE and the
   RESERVE` messages. If, however, the path changes (due to re-routing
   or due to mobility) then an adversary could inject RESERVE` messages
   (with a previously seen SESSION_ID) and could potentially cause harm.

   An off-path adversary can, of course, create RESERVE` messages that
   cause intermediate nodes to create some state (and cause other
   actions) but the message would finally hit the QNE Egress node. The
   QNE Egress node would then be able to determine that there is
   something going wrong and generate an error message.

   The severe congestion handling can be triggered by intermediate nodes
   (unlike other messages). In many cases, however, intermediate nodes
   experiencing congestion use refresh messages modify the <S> and
   <Overload %> parameters of the message. These messages are still
   initiated by the QNE Ingress node and carry the SESSION_ID. The QNE
   Egress node will use the SESSION_ID and subsequently the
   BOUND_SESSION_ID, maintained by the intra-domain QoS-NSLP operational
   state, to refer to a flow that might be terminated. The
   aspect of intermediate nodes initiating messages for severe
   congestion handling is for further study.

   QNE Ingress    QNE Interior   QNE Interior    QNE Egress
 NTLP stateful  NTLP stateless  NTLP stateless  NTLP stateful
        |               |               |              |
        | REFRESH RESERVE`              |              |
        +-------------->| REFRESH RESERVE`             |
        | (+RII)        +-------------->| REFRESH RESERVE`
        |               | (+RII)        +------------->|
        |               |               | (+RII)       |
        |               |               |              |
        |               |               |     REFRESH  |
        |               |               |     RESPONSE`|
        |<---------------------------------------------+
        |               |               |     (+RII)   |

                 Figure 25: RMD REFRESH message exchange

   During the refresh procedure a RESERVE` creates a RESPONSE`, see
   Figure 25. The RII is carried in the RESERVE` message and the
   RESPONSE` message that is generated by the QNE Egress node contains
   the same RII as the RESERVE`.

   The RII can be used by the QNE Ingress to match the RESERVE` with the
   RESPONSE`.

Bader, et al.                                                 [Page 77]


INTERNET-DRAFT                                                 RMD-QOSM

   A further aspect is marking of data traffic. Data packets can be
   modified by an intermediary without any relationship to a signaling
   session (and a SESSION_ID). The problem appears if an off-path
   adversary injects spoofed data packets. The adversary thereby needs
   to spoof data packets that relate to the flow identifier of an
   existing end-to-end reservation that SHOULD be terminated. Therefore
   the question arises how an off-path adversary SHOULD create a data
   packet that matches an existing flow identifier (if a 5-tuple is
   used). Hence, this might not turn out to be simple for an adversary
   unless we assume the previously mentioned mobility/re-routing case
   where the path through the network changes and the set of nodes that
   are along a path changes over time.


6.  IANA Considerations

   A new QOSM ID ("RMD QOSM") needs to be assigned by IANA. The value
   Has to be placed into the QSPEC Type registry that was created with
   [QSP-T].

   The following new parameters are registered in the Parameter ID
   registry that was created with [QSP-T]:

            Value          Parameter/Container ID
   ------------------------+----------------
   |To-be-assigned-by-IANA | PHR Container |
   |To-be-assigned-by-IANA | Bandwidth     |
   |To-be-assigned-by-IANA | PDR Container |
   ------------------------+----------------

   Note to the editor: in this draft a list with temporarily parameter
   ID values are given to the Bandwidth parameter, PHR containers and
   PDR containers, see Sections 4.1.1, 4.1.2 and 4.1.3. The temporarily
   ID value given for the Bandwidth parameter ID is Bandwidth_ID, see
   Section 4.1.1. The given PHR container ID values are in a range from
   PHR_1 to PHR_3 and the given PDR container ID values are in the range
   from PDR_4 to PDR_10. After the IANA will assign new parameter ID
   values, then all these temporarily assigned values have to be
   reassigned.


7.  Acknowledgments

   The authors express their acknowledgement to people who have worked
   on the RMD concept: Z. Turanyi, R. Szabo, G. Pongracz, A. Marquetant,
   O. Pop, V. Rexhepi, G. Heijenk, D. Partain, M. Jacobsson, S.
   Oosthoek, P. Wallentin, P. Goering, A. Stienstra, M. de Kogel, M.
   Zoumaro-Djayoon, M. Swanink, R. Klaver G. Stokkink, J. W. van
   Houwelingen, D. Dimitrova, T. Sealy, H. Chang.


Bader, et al.                                                 [Page 78]


INTERNET-DRAFT                                                 RMD-QOSM

8.  Authors' Addresses

   Attila Bader
   Ericsson Research
   Ericsson Hungary Ltd.
   Laborc 1, Budapest, Hungary, H-1037
   EMail: Attila.Bader@ericsson.com

   Lars Westberg
   Ericsson Research
   Torshamnsgatan 23
   SE-164 80 Stockholm, Sweden
   EMail: Lars.Westberg@ericsson.com

   Georgios Karagiannis
   University of Twente
   P.O.  BOX 217
   7500 AE Enschede, The Netherlands
   EMail: g.karagiannis@ewi.utwente.nl

   Cornelia Kappler
   DeZem GmbH
   Knesebeckstr. 86/87
   10623 Berlin
   Germany
   Email: cornelia.kappler@googlemail.com

   Hannes Tschofenig
   Nokia Siemens Networks
   Linnoitustie 6
   Espoo 02600
   Finland
   EMail: Hannes.Tschofenig@nsn.com
   URI: http://www.tschofenig.priv.at

   Tom Phelan
   Sonus Networks
   250 Apollo Dr.
   Chelmsford, MA USA 01824
   EMail: tphelan@sonusnet.com

   Attila Takacs
   Ericsson Research
   Ericsson Hungary Ltd.
   Laborc 1, Budapest, Hungary, H-1037
   EMail: Attila.Takacs@ericsson.com

   Andras Csaszar
   Ericsson Research
   Ericsson Hungary Ltd.
   Laborc 1, Budapest, Hungary, H-1037
   EMail: Andras.Csaszar@ericsson.com


Bader, et al.                                                 [Page 79]


INTERNET-DRAFT                                                 RMD-QOSM

9.  Normative References

   [RFC2119] S. Bradner, "Key words for use in RFCs to Indicate
   Requirement Levels", RFC 2119, March 1997.

   [QoS-NSLP] Manner, J., Karagiannis, G.,McDonald, A.,
   "NSLP for Quality-of-Service signaling", draft-ietf-nsis-qos-
   nslp (work in progress).

   [QSP-T] Ash, J., Bader, A., Kappler C., "QoS-NSLP QSpec Template"
   draft-ietf-nsis-qspec (work in progress).

   [GIST]  Schulzrinne, H., Hancock, R., "GIST: General Internet
   Messaging Protocol for Signaling", draft-ietf-nsis-ntlp
   (work in progress).

10.  Informative References

   [AdCa03] Adler, M., Cai, J.-Y., Shapiro, J. K., Towsley, D.,
   "Estimation of congestion price using probabilistic packet marking",
   Proc. IEEE INFOCOM, pp. 2068-2078, 2003.

   [AnHa06] Lachlan L. H. Andrew and Stephen V. Hanly, "The Estimation
   Error of Adaptive Deterministic Packet Marking", 44th Annual Allerton
   Conference on Communication, Control and Computing, 2006.

   [AtLi01] Athuraliya, S., Li, V. H., Low, S. H., Yin, Q., "REM: active
   queue management", IEEE Network, vol. 15, pp. 48-53, May/June 2001.

   [Chan07] H. Chang, "Security support in RMD-QOSM", Masters thesis,
   University of Twente, 2007.

   [CsTa05]  Csaszar, A., Takacs, A., Szabo, R., Henk, T., "Resilient
   Reduced-State Resource Reservation", Journal of Communication and
   Networks, Vol. 7, Nr. 4, December 2005.

   [DiKa08]  Dimitrova, D., Karagiannis, G., de Boer, P.-T., "Severe
   congestion handling approaches in NSIS RMD domains with bi-
   directional reservations", Journal of Computer Communications,
   Elsevier, vol. 31, pp. 3153-3162, 2008.

   [JaSh97]  Jamin, S., Shenker, S., Danzig, P., "Comparison of
   Measurement-based Admission Control Algorithms for Controlled-Load
   Service", Proceedings IEEE Infocom `97, Kobe, Japan, April 1997.

   [GrTs03]  Grossglauser, M., Tse, D.N.C, "A Time-Scale Decomposition
   Approach to Measurement-Based Admission Control",  IEEE/ACM
   Transactions on Networking, Vol. 11, No. 4, August 2003

   [RFC3175]  Baker, F., Iturralde, C. Le Faucher, F., Davie, B.,
   "Aggregation of RSVP for IPv4 and IPv6 Reservations",
   IETF RFC 3175, 2001.

Bader, et al.                                                 [Page 80]


INTERNET-DRAFT                                                 RMD-QOSM

   [RFC4125] Le Faucheur & Lai, "Maximum Allocation Bandwidth
   Constraints Model for Diffserv-aware MPLS Traffic Engineering",
   RFC 4125, June 2005.

   [RFC4127] Le Faucheur et al, Russian Dolls Bandwidth Constraints
   Model for Diffserv-aware MPLS Traffic Engineering, RFC 4127, June
   2005

   [RFC1633] Braden R., Clark D., Shenker S., "Integrated Services in
   the Internet Architecture: an Overview", RFC 1633

   [RFC2475] Blake, S., Black, D., Carlson, M., Davies, E., Wang, Z.
   and W.  Weiss, "An Architecture for Differentiated Services", RFC
   2475, December 1998

   [RFC2638] Nichols K., Jacobson V., Zhang L.  "A Two-bit
   Differentiated Services Architecture for the Internet", RFC 2638,
   July 1999

   [RMD1]  Westberg, L., et al., "Resource Management in Diffserv
   (RMD): A Functionality and Performance Behavior Overview", IFIP
   PFHSN`02

   [RMD2] G. Karagiannis, et al., "RMD - a lightweight application
   of NSIS" Networks 2004, Vienna, Austria.

   [RMD3] Marquetant A., Pop O., Szabo R., Dinnyes G., Turanyi Z.,
   "Novel Enhancements to Load Control - A Soft-State, Lightweight
   Admission Control Protocol", Proc. of the 2nd Int. Workshop on
   Quality of Future Internet Services, Coimbra, Portugal,
   Sept 24-26, 2001, pp. 82-96.

   [RMD4] A. Csaszar et al., "Severe congestion handling with
   resource management in diffserv on demand", Networking 2002

   [RFC2998] Bernet Y., Ford, P., Yavatkar, R., Baker, F., Zhang, L.,
   Speer, M., Braden, R., Davie, B., Wroclawski, J. and E.
   Felstaine, "Integrated Services Operation Over Diffserv
   Networks", RFC 2998, November 2000.

   [ThCo04] Thommes, R. W., Coates, M. J., "Deterministic packet marking
   for congestion packet estimation" Proc. IEEE Infocom, 2004.


Bader, et al.                                                 [Page 81]


INTERNET-DRAFT                                                 RMD-QOSM


Appendix A.1.1 Example of a remarking operation during severe
congestion in the Interior nodes

   Per supported PHB, the interior node can support the operation states
   depicted in Figure A.1, when the per-flow congestion notification
   based on probing signaling scheme is used in combination with this
   severe congestion type. Figure A.2 depicts the same functionality
   when the per-flow congestion notification based on probing scheme is
   not used in combination with the severe congestion scheme. The
   description given in this and the following appendices, focuses
   on the situation that during the congestion notification state, the
   "notified DSCP" marking and during the severe congestion state the
   "encoded DSCP" and "affected DSCP" markings are used. In this case,
   the "notified DSCP" marking is used during the
   congestion notification state to mark all packets passing through an
   interior node that operates in the congestion notification state. In
   this way, and in combination with probing, an ECMP solution can be
   provided for the congestion notification state. The "encoded DSCP"
   marking is used to encode and signal the excess rate,
   measured at interior nodes, to the egress nodes. The "affected DSCP"
   marking is used to mark all packets that are passing through a severe
   congested node and are not "encoded DSCP" marked.
   Another possible situation could be derived where both congestion
   notification and severe congestion state are using the "encoded DSCP"
   marking, without using the "notified DSCP" marking. The "affected
   DSCP" marking is used to mark all packets that are passing through
   an Interior node that is in severe congestion state and are not
   "encoded DSCP" marked. In addition, the probe packet that is carried
   by an intra-domain RESERVE message and pass through Interior nodes
   SHOULD be "encoded DSCP" marked if the Interior node is in
   congestion notification or severe congestion states. Otherwise the
   probe packet will remain unmarked. In this way an ECMP solution can
   be provided for both congestion notification and severe congestion
   states. The"encoded DSCP" packets are signaling excess rate that is
   not only associated with interior nodes that are in severe congestion
   state, but also with interior nodes that are in congestion
   notification state.  The algorithm at the Interior node, is similar
   to the algorithm described in the following appendix sections.
   However, this method is not described in detail in this example.

Bader, et al.                                                [Page 82]


INTERNET-DRAFT                                                 RMD-QOSM

                ---------------------------------------------
               |        event B                              |
               |                                             V
            ----------             -------------           ----------
           | Normal   |  event A  | Congestion  | event B | Severe   |
           |  state   |---------->| notification|-------->|congestion|
           |          |           |  state      |         |  state   |
            ----------             -------------           ----------
             ^  ^                       |                     |
             |  |      event C          |                     |
             |   -----------------------                      |
             |         event D                                |
              ------------------------------------------------

        Figure A.1: States of operation, severe congestion combined with
        congestion notification based on probing

            ----------                 -------------
           | Normal   |  event B      | Severe      |
           |  state   |-------------->| congestion  |
           |          |               |  state      |
            ----------                 -------------
                ^                           |
                |      event E              |
                 ---------------------------
        Figure A.2: States of operation, severe congestion without
        congestion notification based on probing

   The terms used in Figure A.1 and Figure A.2 are:

   Normal state: represents the normal operation conditions of the
   node,   i.e. no congestion

   Severe congestion state: it represents the state when state the
   interior node is severely congested related to a certain PHB. It is
   important to emphasize that one of the targets of the severe
   congestion state solution to change the severe congestion state
   behaviour directly to the normal state.

   Congestion notification: state where the load is relatively high,
   close to the level when congestion can occur

   event A: this event occurs when the incoming PHB rate is higher than
   the "congestion notification detection" threshold and lower than the
   severe congestion detection". This threshold is used by the
   congestion notification based on probing scheme, see
   Section 4.6.1.7, 4.6.2.6.

   event B: this event occurs when the incoming PHB rate is higher than
   the "severe congestion detection" threshold.

Bader, et al.                                                [Page 83]


INTERNET-DRAFT                                                 RMD-QOSM

   event C: this event occurs when the incoming PHB rate is lower than
   or equal to the "congestion notification detection" threshold.

   event D: this event occurs when the incoming PHB rate is lower than
   or equal to the "severe_congestion_restoration" threshold. It is
   important to emphasize that this even supports one of the targets of
   the severe congestion state solution to change the severe congestion
   state behaviour directly to the normal state.

   event E: this event occurs when the incoming PHB rate is lower than
   or equal to the "severe congestion restoration" threshold.

   Note that the "severe congestion detection", "severe congestion
   restoration" and admission thresholds SHOULD be higher than the
   "congestion notification detection" threshold, i.e.,:
   "severe congestion detection" > "congestion notification detection"
   and "severe congestion restoration" > "congestion notification
   detection"

   Furthermore, the "severe congestion detection" threshold SHOULD be
   higher than or equal to the admission threshold that is used by the
   reservation based and NSIS measurement based signaling schemes.
   "severe congestion detection" >= admission threshold

   Moreover, the "severe congestion restoration" threshold SHOULD be
   lower than or equal to the "severe congestion detection" threshold
   that is used by the reservation based and NSIS measurement based
   signaling schemes, i.e.,:

   "severe congestion restoration" <= "severe congestion detection"

   During severe congestion the interior node calculates, per traffic
   class (PHB), the incoming rate that is above the "severe congestion
   restoration" threshold, denoted as signaled_overload_rate, in the
   following way:

   * A severe congested interior node SHOULD take into account that
   packets might be dropped. Therefore, before queuing and eventually
   dropping packets, the interior node SHOULD count the total number of
   unmarked and remarked bytes received by the severe congested node,
   denote this number as total_received_bytes. Note that there are
   situations when more than one interior nodes in the same path become
   severe congested. Therefore, any interior node located behind a
   severe congested node MAY receive marked bytes.

Bader, et al.                                                [Page 84]


INTERNET-DRAFT                                                 RMD-QOSM

   When the "severe congestion detection" threshold per PHB is set
   equal to the maximum capacity allocated to one PHB used by the RMD-
   QOSM it means that if the maximum capacity associated to a PHB is
   fully utilized and a packet belonging to this PHB arrives, then it is
   assumed that the interior node will not forward this packet
   downstream. In other words this packet will either be dropped or set
   to another PHB. Furthermore, this also means that after the severe
   congestion situation is solved, then the ongoing flows will be able
   to send their associated packets up to a total rate equal to the
   maximum capacity associated to the PHB. Therefore, when more than one
   interior nodes located on the same path will be severe congested and
   when the interior node receives "encoded DSCP" marked packets, then
   it will mean that an interior node located upstream is also severely
   congested.

   When the "severe congestion detection" threshold per PHB
   is set equal to the maximum capacity allocated to one PHB, then this
   interior node MUST forward the "encoded DSCP" marked packets and it
   SHOULD NOT consider these packets during its local remarking process.
   In other words, the egress should see the excess rates encoded by the
   different severe congested interior nodes as independent, and
   therefore, these independent excess rates will be added.

   When the "severe congestion detection" threshold per PHB
   is not set equal to the maximum capacity allocated to one PHB then
   this means that after the severe congestion situation is solved, the
   ongoing flows will not be able to send their associated packets
   up to a total rate equal to the maximum capacity associated to the
   PHB, but only up to the "severe_congestion_threshold". When more than
   one interior nodes located on the same communication path are severe
   congested and when one of these interior node receives "encoded_DSCP"
   marked packets then this interior node SHOULD NOT mark unmarked,
   i.e., either "original DSCP" or "affected DSCP" or "notified DSCP""
   encoded packets, up to a rate equal to the difference between the
   maximum PHB capacity and the "severe congestion threshold", when the
   incoming "encoded DSCP"" marked packets are already able to signal
   this difference. In this case the "severe congestion threshold"
   SHOULD be configured in all interior nodes, which are located in the
   RMD domain, and equal to:

  "severe_congestion_threshold" =
   = Maximum PHB capacity - threshold_offset_rate

   The threshold_offset_rate represents rate and SHOULD have the same
   value in all interior nodes.

   * before queuing and eventually dropping the packets, at the end of
   each measurement interval of T seconds, calculate the current
   estimated overloaded rate, say measured_overload_rate, by using the
   following equation:

   measured_overload_rate =
   =((total_received_bytes)/T)-severe_congestion_restoration)

Bader, et al.                                                 [Page 85]


INTERNET-DRAFT                                                 RMD-QOSM

   To provide reliable estimation of the encoded information several
   techniques can be used, see [AtLi01], [AdCa03], [ThCo04], [AnHa06].
   Note that since marking is done in interior nodes, the decisions are
   made at egress nodes, and termination of flows are performed by
   ingress nodes, there is a significant delay until the overload
   information is learned by the ingress nodes, see Section 6 of
   [CsTa05]). The delay consists of the trip time of data packets from
   the severe congested interior node to the egress, the measurement
   interval, i.e., T, and the trip time of the notification signaling
   messages from egress to ingress. Moreover, until the overload
   decreases at the severe congested interior node, an additional trip
   time from the ingress node to the severe congested interior node MUST
   expire. This is because immediately before receiving the congestion
   notification, the ingress MAY have sent out packets in the flows that
   were selected for termination. That is, a terminated flow MAY
   contribute to congestion for a time longer that is taken from the
   ingress to the interior node. Without considering the above, interior
   nodes would continue marking the packets until the measured
   utilization falls below the severe congestion restoration threshold.
   In this way, at the end more flows will be terminated than necessary,
   i.e., an over-reaction takes place. [CsTa05] provides a solution to
   this problem, where the interior nodes use a sliding window memory to
   keep track of the signaling overload in a couple of previous
   measurement intervals. At the end of a measurement intervals, T,
   before encoding and signaling the overloaded rate as "encoded DSCP"
   packets, the actual overload is decreased with the sum of already
   signaled overload stored in the sliding window memory, since that
   overload is already being handled in the severe congestion handling
   control loop. The sliding window memory consists of an integer number
   of cells, i.e, n = maximum number of cells. Guidelines for
   configuring the sliding window parameters are given in [CsTa05].

   At the end of each measurement interval, the newest calculated
   overload is pushed into the memory, and the oldest cell is dropped.

   If Mi is the overload_rate stored in ith memory cell (i = [1..n]),
   then at the end of every measurement interval, the overload rate that
   is signaled to the egress node, i.e., signaled_overload_rate is
   calculated as follows:

   Sum_Mi =0
   For i =1 to n
   {
   Sum_Mi = Sum_Mi + Mi
   }

   signaled_overload_rate = measured_overload_rate - Sum_Mi,

   where Sum_Mi is calculated as above.

   Next, the sliding memory is updated as follows:
       for i = 1..(n-1): Mi <- Mi+1
       Mn <- signaled_overload_rate

Bader, et al.                                                [Page 86]


INTERNET-DRAFT                                                 RMD-QOSM

   The bytes that have to be remarked to satisfy the signaled overload
   rate: signaled_remarked_bytes, are calculated using the
   following pseudo code:
   IF severe_congestion_threshold <> Maximum PHB capacity
   THEN
    {
     IF (incoming_encoded-DSCP_rate <> 0) AND
        (incoming_encoded-DSCP_rate =< termination_offset_rate)
     THEN
        { signaled_remarked_bytes =
         = ((signaled_overload_rate - incoming_encoded-DSCP_rate)*T)/N
        }
     ELSE IF (incoming_encoded-DSCP_rate > termination_offset_rate)
     THEN signaled_remarked_bytes =
         = ((signaled_overload_rate - termination_offset_rate)*T)/N
     ELSE IF (incoming_encoded-DSCP_rate =0)
     THEN signaled_remarked_bytes =
         = signaled_overload_rate*T/N
     }
    ELSE signaled_remarked_bytes =  signaled_overload_rate *T/N

    Where the incoming "encoded DSCP" rate is calculated as follows:
    incoming_encoded-DSCP_rate =
     = (received number of "encoded_DSCP" during T) * N)/T;

   The signal_remarked_bytes represents also the number of
   the outgoing packets (after the dropping stage) that MUST be
   remarked, during each measurement interval T, by a node when operates
   in severe congestion mode.

   Note that in order to process an overload situation higher than 100%
   of the maintained severe congestion threshold all the nodes within
   the domain MUST be configured and maintain a scaling parameter, e.g.,
   N used in the above equation, which in combination with the marked
   bytes, e.g., signaled_remarked_bytes, such a high overload situation
   can be calculated and represented. N can be equal or higher than 1.

   Note that when incoming remarked bytes are dropped, the operation of
   the severe congestion algorithm MAY be affected, e.g., the algorithm
   MAY become in certain situations slower. An implementation of the
   algorithm MAY assure as much as possible that the incoming marked
   bytes are not dropped. This could for example be accomplished by
   using different dropping rate thresholds for marked and unmarked
   bytes.

Bader, et al.                                                [Page 87]


INTERNET-DRAFT                                                 RMD-QOSM

   Note that when the "affected DSCP" marking is used by a node
   that is congested due to a a severe congestion situation then all
   the outgoing packets that are not marked (i.e., by using the "encoded
   DSCP") have to be remarked using the "affected DSCP" marking.
   Furthermore, note that when the congestion notification based on
   probing is used in combination with severe congestion, then in
   addition to the possible "encoded DSCP" and "affected DSCP" another
   DSCP for the remarking of the same PHB is used, see Section 4.6.1.7.
   This additional DSCP is denoted in this document as "notified
   DSCP". When an interior node operates
   in the severe congested state, see Figure A.2, and receives "notified
   DSCP" packets, these packets are considered to be unmarked packets
   (but not "affected DSCP" packets). This means that during severe
   congestion also the "notified DSCP" packets can be remarked and
   encoded as either "encoded DSCP" or "affected DSCP" packets.


Appendix A.1.2 Example of a detailed severe congestion operation in the
Egress nodes

   The states of operation in Egress nodes are similar to the ones
   described in  A.1.1. The definition of the events, see below, is
   however different than the definition of the events given in Figure
   A.1 and Figure A.2:

   * event A: when the egress receives a predefined rate of "notified
   DSCP" marked bytes/packets then event_A is activated, see Section
   4.6.1.7 and A.2.2. The predefined rate of "notified DSCP" marked
   bytes is denoted as the congestion notification detection threshold.
   Note this congestion notification detection threshold can also be
   zero, meaning that the event_A is activated when the egress node,
   during an interval T, receives at least one "notified DSCP" packet.

   * event B: this event occurs when the egress receives packets marked
   as either "encoded DSCP" or "affected DSCP" (when "affected DSCP" is
   applied in the whole RMD domain).

   * event C: this event occurs when the rate of incoming
   "notified DSCP" packets decreases below the congestion notification
   detection threshold. In the situation that the congestion
   notification detection threshold is zero, this will mean that event C
   is activated when the egress node, during an interval T, does not
   receive any "notified DSCP" marked packets.

Bader, et al.                                                [Page 88]


INTERNET-DRAFT                                                 RMD-QOSM

   * event D: this event occurs when the egress, during an interval T,
   does not receive packets marked as either "encoded DSCP" or "affected
   DSCP" (when "affected DSCP" is applied in the whole RMD domain).

   event E: this event occurs when the egress, during an interval T,
   does not receive packets marked as either "encoded DSCP" or
   "affected DSCP" (when "affected DSCP" is applied in the whole RMD
   domain)

   An example of the algorithm for calculation of the
   number of flows associated with each priority class that have to be
   terminated is explained by the pseudo-code below.

   First, when the egress operates in the severe congestion state then
   the total amount of remarked bandwidth associated with the PHB
   traffic class, say total_congested_bandwidth, is calculated.
   Note that when the node maintains information about
   each ingress/egress pair aggregate, then the
   total_congested_bandwidth MUST be calculated per ingress/egress pair
   reservation aggregate. This bandwidth represents the severe congested
   bandwidth that SHOULD be terminated. The total_congested_bandwidth
   can be calculated as follows:

   total_congested_bandwidth = N*input_remarked_bytes/T

   Where, input_remarked_bytes represents the number of "encoded DSCP"
   marked bytes that arrive at the egress, during one measurement
   interval T, N is defined as in Section 4.6.1.6.2.1 and A.1.1. The
   term denoted as terminated_bandwidth is a temporal variable
   representing the total bandwidth that have to be terminated,
   belonging to the same PHB traffic class. The
   terminate_flow_bandwidth(priority_class) is the total of bandwidth
   associated with flows of priority class equal to priority_class. The
   parameter priority_class is an integer fulfilling

   0 =< priority_class =< Maximum_priority.

   The calculate_terminate_flows(priority_class) function determines the
   flows for a given priority class and per PHB that has to be
   terminated. This function also calculates the term
   sum_bandwidth_terminate(priority_class), which is the sum of the
   bandwith associated with the flows that will be terminated.
   The constraint of finding the total number of flows that have to
   be terminated is that sum_bandwidth_terminate(priority_class), SHOULD
   be smaller or approximatelly equal to the variable
   terminate_bandwidth(priority_class).

Bader, et al.                                                [Page 89]


INTERNET-DRAFT                                                RMD-QOSM

     terminated_bandwidth = 0;
     priority_class = 0;
     while terminated_bandwidth < total_congested_bandwidth
      {
       terminate_bandwidth(priority_class) =
       = total_congested_bandwidth - terminated_bandwidth
       calculate_terminate_flows(priority_class);
       terminated_bandwidth =
       = sum_bandwidth_terminate(priority_class) + terminated_bandwidth;
       priority_class = priority_class + 1;
      }

   If the egress node maintains ingress/egress pair reservation
   aggregates, then the above algorithm is performed for each
   ingress/egress pair reservation aggregate.


Appendix A.2.1 Example of a detailed remarking admission control
(congestion notification) operation in Interior nodes

   The predefined congestion notification threshold, see Appendix A.1.1,
   is set according to, and usually less than, an engineered bandwidth
   limitation, i.e., admission threshold, based on e.g. agreed Service
   Level Agreement or a capacity limitation of specific links.
   The difference between the congestion notification threshold and the
   engineered bandwidth limitation, i.e., admission threshold, provides
   an interval where the signaling information on resource limitation is
   already sent by a node but the actual resource limitation is not
   reached. This is due to the fact that data packets associated with an
   admitted session have not yet arrived, while allows the admission
   control process available at the egress to interpret the signaling
   information and reject new calls before reaching congestion. Note
   that in the situation when the data rate is higher than the
   preconfigured congestion notification rate, also data packets are
   re-marked, see section 4.6.1.6.2.1. To distinguish between congestion
   notification and severe congestion, two methods MAY be used (see
   Appendix 1.1.1):

   * using different DSCP values (re-marked DSCP values). The remarked
   DSCP that is used for this purpose is denoted as "notified DSCP" in
   this document. When this method is used and when the interior node is
   in "congestion notification" state, see A.1.1, then the node SHOULD
   remark all the unmarked bytes passing through the node using the
   "notified DSCP". Note that this method can only be applied if all
   nodes in RMD domain use the "notified" DSCP marking. In this way,
   also probe packets that will pass through the interior node that
   operates in congestion notification state are being encoded using the
   "notified DSCP" marking.

Bader, et al.                                                 [Page 90]


INTERNET-DRAFT                                                 RMD-QOSM

   * Using the "encoded DSCP" marking for congestion notification and
   severe congestion. This method is not described in detail in this
   example appendix.


Appendix A.2.2 Example of a detailed admission control (congestion
notification) operation in Egress nodes

   The admission control congestion notification procedure can be
   applied only if the egress maintains the ingress/egress pair
   aggregate. When the operation state of the ingress/egress pair
   aggregate is the "congestion notification", see Appendix A.1.2, then
   the implementation of the algorithm depends on how the congestion
   notification situation is notified to the egress. As mentioned in
   Section A.2.1, two methods are used:

   * using the "notified DSCP". During a measurement interval T, the
   egress counts the number of "notified DSCP" marked bytes that belong
   to the same PHB and are associated with the same ingress/egress pair
   aggregate, say input_notified_bytes. We denote the rate as
   incoming_notified_rate.

   * using the "encoded DSCP". In this case, during a measurement
   interval T, the egress measures the input_notified_bytes by counting
   the "encoded DSCP" bytes.

   Below only the detail description of the first method is given.

   The incoming congestion_rate can be then calculated as follows:

   incoming_congestion_rate = input_notified_bytes/T

   If the incoming_congestion_rate is higher than a preconfigured
   congestion notification threshold, then the communication path
   between ingress and egress is considered to be congested. Note that
   the pre-congestion notification threshold can be set to zero. In this
   case the egress node will operate in congestion notification state at
   the moment that it receives at least one "notified DSCP" encoded
   packet.

Bader, et al.                                                [Page 91]


INTERNET-DRAFT                                                RMD-QOSM

   When the egress node operates in "congestion notification" state
   and if the end-to-end RESERVE (probe) arrives at the egress, then
   this request SHOULD be rejected. Note that this is happening
   only when the probe packet is either "notified DSCP" or "encoded
   DSCP" marked. In this way it is ensured that the end-to-end RESERVE
   (probe) packet passed through the node that it is congested. This
   feature is very useful when ECMP based routing is used to detect
   Only flows that are passing through the congested router.

   If such an ingress/egress pair aggregated state is not available when
   the (probe) RESERVE message arrives at the egress, then this request
   is accepted if the DSCP of the packet carrying the RESERVE messsage
   is unmarked. Otherwise (if the packet is either "notified DSCP" or
   "encoded DSCP" marked), it is rejected.


Appendix A.3.1 Example of selecting bi-directional flows for termination
during severe congestion

   When a severe congestion occurs on e.g., in the forward path, and
   when the algorithm terminates flows to solve the severe congestion in
   forward path, then the reserved bandwidth associated with the
   terminated bidirectional flows is also released. Therefore, a careful
   selection of the flows that have to be terminated SHOULD take place.
   A possible method of selecting the flows belonging to the same
   priority type passing through the severe congestion point on a
   unidirectional path can be the following:

   * the egress node SHOULD select, if possible, first unidirectional
   flows instead of bidirectional flows
   * the egress node SHOULD select, if possible, bidirectional flows
   that reserved a relatively small amount of resources on the path
   reversed to the path of congestion.


Appendix A.3.2 Example of a severe congestion solution for bi-
directional flows congested simultaneously on forward and reverse path

   This scenario describes a solution using the combination of the
   severe congestion solutions described in Section 4.6.2.5.2.
   It is considered that the severe congestion occurs simultaneously on
   forward and reverse directions, which MAY affect the same bi-
   directional flows.

Bader, et al.                                                [Page 92]


INTERNET-DRAFT                                                 RMD-QOSM


   When the QNE Edges maintain per-flow intra-domain QoS-NSLP
   operational states then the steps can be the following, see Figure
   A.3. Consider that the egress node selects a number of bi-directional
   flows to be terminated. In this case the egress will send for each
   bi-directional flows a NOTIFY message to ingress. If the Ingress
   receives these NOTIFY messages and its operational state (associated
   with reverse path) is in the severe congestion state (see Figure A.1
   and A.2), then the ingress operates in the following way:

QNE (Ingress)    NE (int.)    NE (int.)       NE (int.)    QNE (Egress)
NTLP stateful                                             NTLP stateful
data|    user        |                |           |               |
--->|    data        | #unmarked bytes|           |               |
    |--------------->S #marked bytes  |           |               |
    |                S--------------------------->|               |
    |                |                |           |-------------->|data
    |                |                |           |               |--->
    |                |                |           |              Term.?
    |            NOTIFY               |           |               |Yes
    |<------------------------------------------------------------|
    |                |                |           |               |data
    |                |                |  user     |               |<---
    |   user data    |                |  data     |<--------------|
    | (#marked bytes)|                S<----------|               |
    |<--------------------------------S           |               |
    | (#unmarked bytes)               S           |               |
Term|<--------------------------------S           |               |
Flow?                |                S           |               |
YES |RESERVE(RMD-QSpec):              S           |               |
    |"forward - T tear"               s           |               |
    |--------------->|  RESERVE(RMD-QSpec):       |               |
    |                |  "forward - T tear"        |               |
    |                |--------------------------->|               |
    |                |                S           |-------------->|
    |                |                S         RESERVE(RMD-QSpec):
    |                |                S       "reverse - T tear"  |
    |      RESERVE(RMD-QSpec)         S           |<--------------|
    |      "reverse - T tear"         S<----------|               |
    |<--------------------------------S           |               |

  Figure A.3: Intra-domain RMD severe congestion handling for
           bi-directional reservation (congestion on both forward and
           reverse direction)

Bader, et al.                                                 [Page 93]


INTERNET-DRAFT                                                 RMD-QOSM

   * For each NOTIFY message, the Ingress SHOULD identify the
   bidirectional flows have to be terminated.

   * The ingress then calculates the total bandwidth that SHOULD be
   released in the reverse direction (thus not in forward direction) if
   the bidirectional flows will be terminated (preempted), say
   "notify_reverse_bandwidth". This bandwidth can be calculated by the
   sum of the bandwidth values associated with all the end-to-end
   sessions that received a (severe congestion) NOTIFY message.

   * Furthermore, using the received marked packets (from the reverse
   path) the ingress will calculate, using the algorithm used by an
   egress and described in A.1.2, the total bandwidth that has to be
   terminated in order to solve the congestion in the reverse path
   direction, say "marked_reverse_bandwidth".

   * The ingress then calculates the bandwidth of the additional flows
   that have to be terminated, say "additional_reverse_bandwidth", in
   order to solve the severe congestion in reverse direction, by taking
   into account:

   ** the bandwidth in the reverse direction of the bidirectional flows
   that were appointed by the egress (the ones that received a NOTIFY
   message) to be preempted, i.e., "notify_reverse_bandwidth"

   **  the total amount of bandwidth in the reverse direction that has
   been calculated by using the received marked packets, i.e.,
   "marked_reverse_bandwidth".
   This additional bandwidth can be calculated using the following
   algorithm:

    IF ("marked_reverse_bandwidth" > "notify_reverse_bandwidth") THEN
    "additional_reverse_bandwidth" =
     = "marked_reverse_bandwidth"- "notify_reverse_bandwidth";
    ELSE
    "additional_reverse_bandwidth" = 0

* Ingress terminates the flows that experienced a severe congestion
in the "forward" path and received a (severe congestion) NOTIFY
   message

   * If possible the ingress SHOULD terminate unidirectional flows that
   are using the same egress-ingress reverse direction communication
   path to satisfy the release of a total bandiwtdh up equal to the:
   "additional_reverse_bandwidth", see Appendix 3.1.

Bader, et al.                                                 [Page 94]


INTERNET-DRAFT                                                 RMD-QOSM

   * If the number of REQUIRED uni-directional flows (to satisfy the
   above issue) is not available, then a number of bi-directional flows
   that are using the same egress-ingress reverse direction
   communication path MAY be selected for pre-emption in order to
   satisfy the release of a total bandiwtdh equal up to the:
   "additional_reverse_bandwidth".  Note that using the guidelines given
   in Appendix A.3.1, first the bidirectional flows that reserved a
   relatively small amount of resources on the path reversed to the path
   of congestion SHOULD be selected for termination.

   When the QNE Edges maintain aggregated intra-domain QoS-NSLP
   operational states then the steps can be the following.

   * The egress calculates the bandwidth to be terminated using the same
   method as described in Section 4.6.1.6.2.2. The egress includes this
   bandwidth value in a <PDR Bandwidth> within a "PDR_Congestion_Report"
   container that is carried by the end-to-end NOTIFY message.

   * The Ingress receives the NOTIFY message and reads the <PDR
   Bandwidth> value included in the "PDR_Congestion_Report" container.
   Note that this value is denoted as "notify_reverse_bandwidth" in the
   situation that the QNE edges maintain per flow intra-domain QoS-NSLP
   operational states, but is calculated differently. The variables
   "marked_reverse_bandwidth" and "additional_reverse_bandwidth are
   calculated using the same steps as explained for the situation that
   the QNE edges maintain per flow intra-domain QoS-NSLP states.

   * Regarding the termination of flows that are using the same egress-
   ingress reverse direction communication path, the Ingress can follow
   the same procedures as the situation that the QNE edges
   maintain per-flow intra-domain QoS-NSLP operational states.

   The RMD aggregated (reduced state) reservations maintained by the
   interior nodes, can be reduced in the "forward" and "reverse"
   directions by using the procedure described in Section 4.6.2.3 and
   including in the  <Bandwith> parameter within the "RMD-QOSM QOS
   Description" field carried by the "forward" intra-domain RESERVE the
   value equal to "notify_reverse_bandwidth" and by including the
   "additional_reverse_bandwidth" value in the <PDR Bandwidth> parameter
   within the "PDR_Release_Request" container that is carried by the
   same intra-domain RESERVE message.


Bader, et al.                                                 [Page 95]


INTERNET-DRAFT                                                 RMD-QOSM


Appendix A.4 Pre-emption handling during admission control

   This section describes the mechanism that can be supported by the
   QNE Ingress, QNE Interior and QNE Egress nodes to satisfy
   pre-emption during the admission control process.
   This mechanism uses the pre-emption building blocks specified in
   [QoS-NSLP].

A.4.1 Pre-emption handling in QNE Ingress nodes

   If a QNE Ingress receives a RESERVE for a session that
   causes other session(s) to be pre-empted, for each of these to be
   pre-empted sessions, then the QNE Ingress follows the following
   steps:

   Step_1:

   The QNE Ingress MUST send a tearing RESERVE downstream and add a
   BOUND_SESSION_ID, with Binding_Code value equal to "Indicated session
   caused pre-emption" that indicates the SESSION_ID of the session that
   caused the pre-emption. Furthermore, an INFO-SPEC object with error
   code value equal to "Reservation pre-empted" has to be included in
   each of these tearing RESERVE messages.
   The selection of which flows have to be preempted can be based on
   predefined policies. For example, this selection process can be based
   on the MRI associated with the high and low priority sessions. In
   particular, the QNE Ingress can select low(er) priority session(s)
   where their MRI is "close" (especially the target IP) to the one
   associated with the higher priority session. This means that
   typically the high priority session and the to be preempted lower
   priority sessions are following the same communication path and are
   passing through the same QNE Egress node.

   Furthermore, the amount of lower priority sessions that have to be
   pre-empted per each high priority session, has to be such that the
   requested resources by the higher priority session SHOULD be lower or
   equal than the sum of the reserved resources associated with the
   lower priority sessions that have to be pre-empted.

   Step_2:

   For each of the sent tearing RESERVE(s) the QNE Ingress will send a
   NOTIFY message with an INFO-SPEC objects with error code value equal
   to "Reservation pre-empted" towards the QNI.

   Step_3:

   After sending the pre-empted (tearing) RESERVE(s), the Ingress QNE
   will send the (reserving) RESERVE, which caused the pre-emption,
   downstream towards the QNE Egress.


Bader, et al.                                                 [Page 96]


INTERNET-DRAFT                                                 RMD-QOSM

A.4.2 Pre-emption handling in QNE Interior nodes

   The QNE Interior upon receiving the first (tearing) RESERVE that
   carries the BOUND_SESSION_ID object with Binding_Code value
   equal to "Indicated session caused pre-emption" and an INFO-SPEC
   object with error code value equal to "Reservation
   preempted" it considers that this session has to be pre-empted.
   In this case the QNE Interior creates a so called "pre-emption
   state", which is identified by the SESSION_ID carried in the
   pre-emption related BOUND_SESSION_ID object. Furthermore, this
   "pre-emption state" will include the SESSION_ID of the session
   associated with the (tearing) RESERVE. If subsequently additional
   tearing RESERVE(s) are arriving including the same values of
   BOUND_SESSION_ID and INFO-SPEC objects, then the associated
   SESSION_IDs of these (tearing) RESERVE message will be included in
   the already created "pre-emption state". The QNE will then set a
   timer, with a value that is high enough to ensure that it will not
   expire before the (reserving) RESERVE arrives.
   Note that when the "pre-emption state" timer expires then the
   bandwidth associated with the pre-empted session(s) will have to be
   released, following a normal RMD-QOSM bandwidth release procedure..
   If the QNE interior node will not receive the all to be pre-empted
   (tearing) RESERVE messages sent by the QNE Ingress before their
   associated (reserving) RESERVE message arrives, then the (reserving)
   RESERVE message will not reserve any resources and this message will
   be "M" marked, see Section 4.6.1.2. Note that this situation is not a
   typical situation. Typically, this situation can only occur when at
   least one of (tearing) RESERVe messages are dropped due to an error
   condition.

   Otherwise, if the QNE Interior receives the all to be pre-empted
   (tearing) RESERVE messages sent by the QNE Ingress, then the QNE
   Interior will remove the pending resources, and make the new
   reservation using normal RMD-QOSM bandwidth release and reservation
   procedures.


A.4.3 Pre-emption handling in QNE Egress nodes

   Similar to the QNE Interior operation, the QNE Egress upon receiving
   the first (tearing) RESERVE that carries the BOUND_SESSION_ID object
   with Binding_Code value equal to "Indicated session caused
   pre-emption" and an INFO-SPEC object with error code value equal to
   "Reservation preempted" it considers that this session has to be pre-
   empted. Similar to the QNE Interior operation the QNE Egress creates
   a so called "pre-emption state", which is identified by the
   SESSION_ID carried in the pre-emption related BOUND_SESSION_ID
   object. This "pre-emption state" will store the same type of
   information and use the
   same timer value as specified in section A.4.2.


Bader, et al.                                                 [Page 97]


INTERNET-DRAFT                                                 RMD-QOSM


   If subsequently additional tearing RESERVE(s) are
   arriving including the same values of BOUND_SESSION_ID and INFO-SPEC
   objects, then the associated SESSION_IDs of these (tearing) RESERVE
   message will be included in the already created "pre-emption state".
   If the (reserving) RESERVE sent by the QNE Ingress node arrived and
   is not "M" marked and if all the to be pre-empted (tearing) RESERVE
   messages arrived then the QNE Egress will remove the pending
   resources and make the new reservation using normal RMD-QOSM
   procedures.

   If the QNE Egress receives a "M" marked RESERVE message then the QNE
   Egress will use the normal partial RMD-QOSM procedure to release the
   partial reserved resources associated with the "M" marked RESERVE,
   see Section 4.6.1.2.

   If the QNE Egress will not receive all the to be pre-empted (tearing)
   RESERVE messages sent by the QNE Ingress before their associated and
   not "M" marked (reserving) RESERVE message arrives, then the
   following steps can be followed:

 * If the QNE Egress uses an end-to-end QOSM supports the pre-emption
   handling then the QNE Egress have to calculate and select new
   lower priority sessions that have to be terminated. The way of how
   the to be pre-empted sessions are selected and signalled to the
   downstream QNEs is similar to the operation specified in Section
   A.4.1.
 * If the QNE Egress does not use an end-to-end QOSM that supports
   the pre-emption handling then the QNE Egress has to reject the
   requesting (reserving) RESERVE associated with the high priority
   session, see Section 4.6.1.2.

   Note that typically, the situation that the QNE Egress does not
   receive all the to be pre-empted (tearing) RESERVE messages sent by
   the QNE Ingress can only occur when at least one of (tearing) RESERVe
   messages are dropped due to an error condition.


A.5 Retransmission procedure within the RMD domain

   If the retransmission of intra-domain RESERVE messages within the RMD
   domain is not disallowed then all the QNE Interior nodes SHOULD use
   the functionality described in this section.

   In this situation we enable QNE Interior nodes to maintain a replay
   cache in which each entry contains the <RSN>, <SESSION_ID> (available
   via GIST), <REFRESH_PERIOD> (available via the QoS NSLP [QoS-NSLP])
   and the last received "PHR Container" <Container ID> carried by the
   RMD-QSpec for each session [QSP-T]. Thus this solution uses
   information carried by QOS-NSLP objects [QoS-NSLP] and parameters
   carried by the RMD-QSpec "PHR Container". The following phases can be
   distinguished:


Bader, et al.                                                 [Page 98]


INTERNET-DRAFT                                                 RMD-QOSM


   Phase 1: Create Replay Cache Entry

   When an Interior node receives an intra-domain RESERVE
   message and its cache is empty or there is no matching entry, it
   reads the <Container ID> field of the "PHR container" of the received
   message. If the <Container ID> is a PHR_RESOURCE_REQUEST, which
   indicates that the intra-domain Reserve message is a reservation
   request, then the QNE Interior node creates a new entry in the cache
   and copies the <RSN>, <SESSION_ID> and <Container ID> to the entry
   and sets the <REFRESH_PERIOD>.
   By using the information stored in the List the Interior node
   verifies whether the received intra-domain RESERVE message is sent
   by an adversary or not. For example, if the
   <SESSION_ID> and <RSN> of a received intra-domain RESERVE message
   match the values stored in the List then the Interior node checks the
   <Container ID> part.

   If the <Container_ID> is different then:

     Situation D1: <Container ID> in its own list is
        PHR_RESOURCE_REQUEST, and <Container ID> in the message is
        PHR_REFRESH_UPDATE;

     Situation D2: <Container ID> in its own list is
        PHR_RESOURCE_REQUEST or PHR_REFRESH_UPDATE, and <Container ID>
        in the message is PHR_RELEASE_REQUEST;

     Situation D3: <Container ID> in its own list is PHR_REFRESH_UPDATE,
        and <Container ID> in the message is PHR_RESOURCE_REQUEST;

    For Situation D1, the QNE Interior node processes this message by
    RMD-QoSM default operation, reserves bandwidth, updates the entry
    and passes the message to downstream nodes. For Situation D2, the
    QNE Interior node processes this message by RMD-QoSM default
    operation, releases bandwidth, deletes all entries associated with
    the session and passes the message to downstream nodes. For
    situation D3, the QNE Interior node does not use/process the
    <Bandwidth> parameter carried by the received intra-domain RESERVE
    message. Furthermore, the <K> flag in the "PHR Container"
    have to be set such that the <Bandwidth> parameter carried by the
    intra-domain RESERVE message is not processed/used by a QNE
    Interior node.

    If the <Container_ID> is the same then:

     Situation S1: <Container ID> is equal to PHR_RESOURCE_REQUEST;
     Situation S2: <Container ID> is equal to PHR_REFRESH_UPDATE;

Bader, et al.                                                 [Page 99]


INTERNET-DRAFT                                                 RMD-QOSM


     For situation S1, the QNE Interior node does not process the
     intra-domain RESERVE message but it just passes it to
     downstream nodes, because
     it might have been retransmitted by the QNE Ingress node;
     For situation S2, the QNE Interior node processes the first
     incoming intra-domain (refresh) RESERVE message within a refresh
     period and updates the entry and forwards it to the downstream
     nodes.

    If only <Session_ID> is matched to the list, then the QNE Interior
    Node checks the <RSN>. Here also two situations can be
    distinguished:

    If a rerouting takes place, see Section 5.2.5.2 in [QoS-NSLP], the
    <RSN> in the message will be equal either to <RSN + 2> in the
    stored list if it is not a tearing RESERVE or to <RSN -1> in the
     stored list if it is a tearing RESERVE:

     The QNE Interior node will check the <Container ID> part;

     If the <RSN> in the message is equal to <RSN + 2> in the
     stored list and the <Container ID> is a PHR_RESOURCE_REQUEST or
     PHR_REFRESH_UPDATE then the received intra-domain RESERVE
     message has to be interpreted and processed as a typical
     (non tearing) RESERVE message, which is caused by rerouting, see
      Section 5.2.5.2 in [QoS-NSLP].

     If the <RSN> in the message is equal to <RSN-1> in the
     stored list and the <Container ID> is a PHR_RELEASE_REQUEST
     then the received intra-domain RESERVE message has to be
     interpreted and processed as a typical (tearing) RESERVE
     message, which is caused by rerouting, see Section 5.2.5.2 in
     [QoS-NSLP].

     If other situations occur than the ones described above, then
     the QNE Interior node does not use/process the
     <Bandwidth> parameter carried by the received intra-domain
     RESERVE message. Furthermore, the <K> paramer has to be set, see
     above.


    Phase 2: Update Replay Cache Entry

    When a QNE Interior node receives an intra-domain RESERVE message,
    it retrieves the corresponding entry from the cache and compares the
    values. If the message is valid, the Interior node will update
    <Container ID> and <REFRESH_PERIOD> in the List entry.

Bader, et al.                                                [Page 100]


INTERNET-DRAFT                                                 RMD-QOSM


    Phase 3: Delete Replay Cache Entry

    When a QNE Interior node receives an intra-domain (tear) RESERVE
    message and an entry in the replay cache can be found, then the
    QNE Interior node will delete this entry after processing the
    message. Furthermore, the Interior node will delete cache entries,
    if it did not receive an intra-domain (refresh) Reserve message
    during the <REFRESH_PERIOD> period with a <Container_ID> value equal
    to PHR_REFRESH_UPDATE.


Disclaimer

   This document may contain material from IETF Documents or IETF
   Contributions published or made publicly available before November
   10, 2008.  The person(s) controlling the copyright in some of this
   material may not have granted the IETF Trust the right to allow
   modifications of such material outside the IETF Standards Process.
   Without obtaining an adequate license from the person(s)
   controlling the copyright in such materials, this document may not
   be modified outside the IETF Standards Process, and derivative
   works of it may not be created outside the IETF Standards Process,
   except to format it for publication as an RFC or to translate it
   into languages other than English.


Bader, et al.                                                [Page 101]