Network Working Group Gabor Feher, BUTE
INTERNET-DRAFT Istvan Cselenyi, TRAB
Expiration Date: May 2002 Andras Korn, BUTE
November 2001
Benchmarking Terminology for Routers Supporting Resource Reservation
<draft-ietf-bmwg-benchres-term-01.txt>
1. Status of this Memo
This document is an Internet-Draft and is in full conformance with
all provisions of Section 10 of RFC2026.
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 memo provides information for the Internet community. This memo
does not specify an Internet standard of any kind. Distribution of
this memo is unlimited.
2. Table of contents
1. Status of this Memo.............................................1
2. Table of contents...............................................1
3. Abstract........................................................2
4. Introduction....................................................2
5. Existing definitions............................................3
6. Definition of Terms.............................................3
6.1 Resource Reservation Protocol Basics........................3
6.1.1 Resource Reservation Session...........................3
6.1.2 Multicast Resource Reservation Session.................4
6.1.3 Resource Reservation Capable Router....................4
6.1.4 Signaling End-point....................................5
6.1.5 Reservation Initiator..................................5
6.1.6 Reservation Session Maintenance........................6
6.1.7 Signaling Path.........................................7
6.2 Traffic Types...............................................8
6.2.1 Premium Traffic........................................8
Feher, Cselenyi, Korn Expires May 2002 [Page 1]
INTERNET-DRAFT <draft-ietf-bmwg-benchres-term-01.txt> November 2001
6.2.2 Best-Effort Traffic....................................8
6.3 Router Load Types...........................................9
6.3.1 Traffic Load...........................................9
6.3.2 Session Load...........................................9
6.3.3 Signaling Load........................................10
6.3.4 Signaling Burst.......................................11
6.4 Performance Metrics........................................11
6.4.1 Signaling Message Handling Time.......................11
6.4.2 Premium Traffic Delay.................................12
6.4.3 Best-effort Traffic Delay.............................13
6.4.4 Signaling Message Loss................................13
6.4.5 Session Refreshing Capacity...........................14
6.4.6 Scalability Limit.....................................14
7. Acknowledgement................................................15
8. References.....................................................15
9. Authors' Addresses:............................................15
3. Abstract
The purpose of this document is to define terminology specific to the
performance benchmarking of the resource reservation signaling of IP
routers. These terms are used in additional documents that define
benchmarking methodologies for routers supporting resource
reservation and define reporting formats for the benchmarking
measurements.
4. Introduction
The IntServ over DiffServ framework [1] outlines a heterogeneous
Quality of Service (QoS) architecture for multi domain Internet
services. Signaling based resource reservation (e.g. via RSVP [2]) is
an integral part of that model. While this significantly lightens the
load on most of the core routers, the performance of border routers
that handle the QoS signaling is still crucial. Therefore network
operators, who are planning to deploy this model, shall scrutinize
the scalability limitations in reservation capable routers and the
impact of signaling on the forwarding performance of the routers.
An objective way for quantifying the scalability constraints of QoS
signaling is to perform measurements on routers that are capable of
resource reservation. This document defines terminology for specific
set of tests that vendors or network operators can use to measure and
report the signaling performance characteristics of router devices
that support resource reservation protocols. The results of these
tests provide comparable data for different products supporting the
decision process before purchase. Moreover, these measurements
provide input characteristics for the dimensioning of a network in
which resources are provisioned dynamically by signaling. Finally,
the tests are applicable for characterizing the impact of the control
plane signaling on the forwarding performance of routers.
This benchmarking terminology document is based on the knowledge
gained by examination of (and experimentation with) several very
Feher, Cselenyi, Korn Expires May 2002 [Page 2]
INTERNET-DRAFT <draft-ietf-bmwg-benchres-term-01.txt> November 2001
different resource reservation protocols: RSVP [2], Boomerang [5],
YESSIR [6], ST2+ [7], SDP [8] and Ticket [9]. Nevertheless, this
document aspires to compose terms that are valid in general and not
restricted to these protocols.
5. Existing definitions
RFC 1242 [3] "Benchmarking Terminology for Network Interconnect
Devices" and RFC 2285 [4] "Benchmarking Terminology for LAN Switching
Devices" contains discussions and definitions for a number of terms
relevant to the benchmarking of signaling performance of reservation
capable routers and should be consulted before attempting to make use
of this document.
For the sake of clarity and continuity this document adopts the
template for definitions set out in Section 2 of RFC 1242.
Definitions are indexed and grouped together in sections for ease of
reference.
6. Definition of Terms
6.1 Resource Reservation Protocol Basics
This group of definitions applies to various signaling based resource
reservation protocols implemented on IP router devices.
6.1.1 Resource Reservation Session
Definition:
A resource reservation session (or shortly reservation) expresses
that routers along the data path between two network nodes apply
special QoS treatment to a certain traffic flow.
Discussion:
The QoS treatment is specified by giving the amount of networking
resources that are dedicated to the traffic flow during the length
of the reservation session. Depending on the protocol, there are
different approaches to define the network resource requirement of
a traffic flow. It can be described by high-level parameters, like
the required bandwidth, service class or the maximum traffic
delay; or it can be low-level information, like the parameters of
a leaky-bucket model of the traffic flow [10].
Issues:
There are resource reservation protocols, where resource
dedications in a router are unique for each resource reservation
session. However, in this case the number of resource dedications
grows along with the number of sessions and working with huge
number of resource dedications raise problems (see Reservation
Session Maintenance). Therefore, many resource reservation
protocols allow to bunch different reservation sessions into one
aggregated session, which takes only one aggregated resource
allocation for the whole bunch. The aggregation can be based on
Feher, Cselenyi, Korn Expires May 2002 [Page 3]
INTERNET-DRAFT <draft-ietf-bmwg-benchres-term-01.txt> November 2001
the similar attributes of the flows, (e.g. aggregation using
DiffServ code-points [11]) or it can combine arbitrary sessions as
well.
See Also:
Reservation Session Maintenance
6.1.2 Multicast Resource Reservation Session
Definition:
A multicast resource reservation session (or, in short, multicast
reservation) denotes that certain QoS treatment is applied to the
packets of every traffic flow related to a multicast group.
Discussion:
Usually, there are several traffic sources and destinations in a
multicast group. In order to be able to guarantee the QoS
parameters for each packet of the multicast flow, every router
that forwards the multicast traffic must dedicate resources to the
flow.
Generally, there are two types of multicast resource reservations:
many-to-many and one-to-many multicast reservations. Those of the
first type allow traffic to be originated from several sources,
while those of the second type permit only one traffic source in
the whole multicast group and this source should not change during
the lifetime of the session. Additionally, in several cases, a
many-to-many multicast reservation session does not require the
same amount of resources reserved in every involved router.
Depending on the resource reservation protocol, the traffic
destinations of the multicast group may request different QoS
parameters. Furthermore, the protocols may describe more than one
reservation styles expressing the resource reservation
distribution method among the involved routers. (e.g. RSVP
SE/WF/FF [2])
Issues:
Naturally, many-to-many multicast reservation capable protocols
are bound to be more complex than one-to-many or non-multicast
protocols. Usually, the router has to be aware of the location of
the traffic sources and destinations participating in the
multicast reservation in the aspect of its network interfaces,
plus the resource requirements of the traffic destinations in
order to be able to calculate the right amount of resources
dedicated to the session.
6.1.3 Resource Reservation Capable Router
Definition:
By definition, a router is resource reservation capable - supports
resource reservation - if it understands a resource reservation
protocol that signals the set-up, tear-down and modification of
resource reservation sessions.
Feher, Cselenyi, Korn Expires May 2002 [Page 4]
INTERNET-DRAFT <draft-ietf-bmwg-benchres-term-01.txt> November 2001
Discussion:
Resource reservation protocols define signaling messages that are
interpreted by resource reservation capable routers. The router
captures the signaling message and manipulates resource
reservation sessions according to the content of the message. In
addition, the protocol might declare to forward the same or a
modified signaling message to other routers as well.
Issues:
There are resource reservation protocols where routers are
required to initiate signaling messages besides the signaling
message forwarding. The benefits of such protocols are that
changes in resource reservation sessions can be signaled to other
routers immediately, even if the change is not caused by signaling
messages directly. In contrast, the message initiation takes time
that slows down the signaling message processing, so there are
protocols declaring instant responses, where all the signaling
messages are forwarded to other routers immediately, avoiding the
signaling message initiation.
6.1.4 Signaling End-point
Definition:
A signaling end-point is a network node capable of initiating and
terminating resource reservation sessions.
Discussion:
Typically, signaling end-points have a separate protocol stack
that is capable of generating and understanding the signaling
messages. However, in some special cases, the resource reservation
initiation is carried out without the notice of the signaling
terminating node. For example, the Boomerang resource reservation
protocol encapsulates the reservation requests in an ICMP Echo
message. This message is bounced back from the signaling
terminating network node and as a result the node becomes a
signaling end-point without understanding the reservation
protocol.
There are reservation gateways that translate the signaling
messages of one resource reservation protocol into messages of
another resource reservation protocol. Thus the reservation
gateway represents two signaling end-points in one, as it is both
a signaling terminator and a signaling initiator.
6.1.5 Reservation Initiator
Definition:
The reservation initiator is the signaling end-point that
initiates the resource reservation session setup.
Discussion:
Feher, Cselenyi, Korn Expires May 2002 [Page 5]
INTERNET-DRAFT <draft-ietf-bmwg-benchres-term-01.txt> November 2001
Resource reservation protocols can be classified depending on the
relationship between the reservation initiators and their role in
the traffic flow.
In the case of receiver-oriented protocols, the traffic
destinations, which are the receivers of the data traffic,
initiate the reservation session setup, unlike the sender-oriented
protocols where this is done by traffic sources. Moreover, there
are protocols where both the traffic source and destination can
act as the reservation initiator.
The importance of the reservation initiator orientation is only
dominant in case of multicast reservation sessions. Generally, in
multicast groups the number of traffic destinations changes more
frequently than the number of traffic sources. In this case, the
receiver-oriented protocols do not require the traffic sources to
change their states generating signaling messages when a new
traffic destination joins or an existing one leaves the group,
since the reservation session is managed by the traffic
destinations.
Issues:
Receiver-oriented resource reservation protocols often require two
pass to setup the reservation session. Since the resource
dedication should take place on all the routers that are on the
path from the sources to the destinations, thus the reservation
initiator must have a method to reach every such router. In the
case of the sender-oriented protocols this method is assured by
sending the signaling traffic same way as the data traffic, which
guarantees that signaling messages goes through the same routers
as the data traffic. However, in the case of the receiver-oriented
protocols the reservation initiation requests go from the
destinations to the sources, which is the opposite direction
compared to the data traffic. Thus, receiver-oriented protocols
must provide a mechanism that at first discovers all the routers
along the path of the data traffic (RSVP PATH message [2]), before
the second pass, where the traffic destinations are ready to
initiate resource reservation requests.
6.1.6 Reservation Session Maintenance
Definition:
Soft-state resource reservation protocols require the routers to
maintain the resource reservation sessions from the initiation
until the teardown of the session. This maintenance often involves
the regular checking and refreshing of the session.
Discussion:
Based on the approach of reservation session maintenance, resource
reservation protocols can be divided into two categories: soft-
state protocols and hard-state protocols.
Feher, Cselenyi, Korn Expires May 2002 [Page 6]
INTERNET-DRAFT <draft-ietf-bmwg-benchres-term-01.txt> November 2001
In the case of hard-state protocols (e.g. ST2 [7]), the resource
reservation session established by a set-up signaling primitive is
permanent and is cancelled only when the corresponding tear-down
signaling primitive arrives to the router. Contrary, in the case
of the soft-state protocols there are no permanent resource
reservations. The resource reservation session have to be
regularly refreshed by appropriate signaling messages. No refresh
signaling message arrived during a certain period is assumed as
the indication that the resource reservation session is not
maintained by the signaling end-points any longer. In such case,
the router tears the session down waiting for no explicit request.
For this reason, soft-state protocols exhibit more robust behavior
than hard-state protocols, since no failures can cause permanent
resource stuck in the routers.
Issues:
Although soft-state protocols are more robust than hard-state
protocols, the frequent processing of refresh signaling messages
might cause serious increase in the router load. Moreover, session
maintenance mechanisms often use timers to watch the refresh
period expirations of the sessions. The maintenance of such timers
and the actions due to the expiration of such timers also
contributes to the router load.
In order to reduce the large number of refresh signaling message
processing overhead, the resource reservation protocol may support
various mechanisms to pack several refresh signaling messages into
one signaling message.
6.1.7 Signaling Path
Definition:
A signaling path is a sequence of network nodes and links along
which signaling messages travel from one signaling end-point to
the other.
Discussion:
The resource reservation protocol must provide that each router,
which is responsible to handle a signaling message, truly receives
the signaling message. Usually it is assured by passing through
the signaling messages along the signaling path, which involves
every router affected by the resource reservation session.
Resource reservation protocol must be also prepared that there are
routers forwarding the data traffic of a resource reservation
session that do not support the actual resource reservation
protocol. In this case the signaling traffic must be tunneled
through the zones of the routers that can not interpret the
signaling messages in order to keep the continuity of the
signaling path.
Issues:
Feher, Cselenyi, Korn Expires May 2002 [Page 7]
INTERNET-DRAFT <draft-ietf-bmwg-benchres-term-01.txt> November 2001
It is not unusual for routers to change their routing from time to
time. One reason for the change can be a failure of a neighboring
router or link. In case of route changes the data traffic will be
forwarded along a different path than the signaling messages used
in establishing the resource dedications for the reservation
session. In order to properly handle this situation, hard-state
protocols have to be much more sophisticated in order to detect
the route change and to re-reserve the resources on the new path.
However, soft-state protocols do not have to worry about such
situation, since the refresh messages can be used to set up the
reservation on the new path and the dedicated resources will
eventually disappear from routers of the obsoleted path.
6.2 Traffic Types
This group of definitions defines traffic types forwarded by resource
reservation capable routers.
6.2.1 Premium Traffic
Definition:
Premium traffic is a traffic type that the router distinguishes
from best-effort traffic (to be defined later) and forwards its
packets according to a QoS agreement.
Discussion:
Traffic that corresponds to a resource reservation session in the
router is premium traffic. The QoS treatment is defined in the
associated flow descriptor that is established by the signaling
messages during the reservation session setup.
The router may distinguish several types of premium traffic (e.g.
delay sensitive traffic, loss sensitive traffic, etc.). Different
types of premium traffic may receive different QoS treatment.
Issues:
The router has to identify every packet whether it has dedicated
resource or not. This can be done by either multi-field
classification [11] using the IP 5-tuple or behavior-aggregate
classification using the DSCP field. However, if a packet claims
that it has an associated resource reservation session in the
router, the router has to find the flow descriptor, which might be
time consuming in routers with vast amounts of resource
reservation sessions.
6.2.2 Best-Effort Traffic
Definition:
Best-effort traffic is a traffic type that has no reservation
entry in the router.
Discussion:
Feher, Cselenyi, Korn Expires May 2002 [Page 8]
INTERNET-DRAFT <draft-ietf-bmwg-benchres-term-01.txt> November 2001
Traffic flows that do not require QoS guarantees along their path
are considered to be best-effort traffic. "Best-effort" means that
the router makes its best effort to forward every data packet, but
does not guarantee anything. This is the most common type of
traffic on today's Internet. (There may be scenarios where
resource reservation is done for BE traffic too, but those are
outside of the focus of this memo.)
6.3 Router Load Types
This group of definitions describes different load component types
that impact only a specific part of the resource reservation capable
router. Categorizing the router load is crucial, since the
conventional router load metric expressing the processing power
utilization of the router does not characterize perfectly the
resource reservation capable router. In the case of routers
supporting resource reservations it is also important to know the
source of the processing power utilization.
6.3.1 Traffic Load
Definition:
Traffic load is the load that is raised by forwarding data traffic
on the router.
Discussion:
It is obvious that forwarding the data packets, which requires
obtaining the routing information and transferring the data packet
between network interfaces, requires processing power. Speaking of
general router measurements only this type of load is considered
as the source of the processing power utilization expressed by the
router load metric. Although the traffic load is the dominant load
component, benchmarking routers supporting resource reservations
must consider other load components also in line with the resource
reservation handling.
Measurement unit:
The amount of the traffic load is represented by the volume of the
data traffic. The volume is measured with the transferred bits
during a specified time unit, which is typically bit per seconds
(bps).
6.3.2 Session Load
Definition:
Session load is the load that manifests itself as the excess
processing power required to keep track of reservation sessions.
Discussion:
All signaling based resource reservation protocol implementation
employ a packet classifier algorithm that distinguishes the flows
having reservations in the router from the others that do not.
Therefore each implementation maintains a list of reservation
Feher, Cselenyi, Korn Expires May 2002 [Page 9]
INTERNET-DRAFT <draft-ietf-bmwg-benchres-term-01.txt> November 2001
session descriptors that is instrumental in keeping track of the
resource reservation dedication. Obviously, the more reservation
sessions are set up on the router, the more complex traffic
classification becomes, and the more time it takes for the
classification algorithm to identify the session descriptor list.
Moreover, in most protocols, not only the traffic flows, but also
signaling messages that manipulate resource reservations on the
router have to identify themselves first, before taking any other
actions. This kind of classification gives extra work for the
router.
Session load also involves the duties related to reservation
session maintenance. The maintenance of timers that watchdog the
reservation session refreshes and the signaling of the reservation
session refresh may cause severe load on the router. Based on the
initiating point of the refresh messages, resource reservation
protocols can be divided into two groups. First, there are
protocols where it is the responsibility of the signaling end-
points to initiate refresh messages, which messages are forwarded
by the routers along the signaling path refreshing the
corresponding session. Second, there are other protocols, where
the session refresh happens between the two peering network nodes
from the signaling path only. In this latter case, the routers and
signaling end-points have their own schedule for the refresh
message initiation. The first approach lightens the load of the
session maintenance task; however, the second approach bears the
ability to adjust the signaling message traffic intensity along
the signaling path.
Measurement unit:
The session load is represented by the number of reservation
sessions in the router.
6.3.3 Signaling Load
Definition:
Signaling load is the load that manifests itself as the time
required to process the incoming signaling messages.
Discussion:
The processing of signaling messages requires processing power
that raises load on the control plane of the router. In the case
of routers where the control plane and the data plane are not
totally independent (e.g. certain parts of the tasks are served by
the same processor; or the architecture has common memory buffers
or transfer buses) the signaling load can have an impact on the
router's packet forwarding performance as well.
Most of the resource reservation protocols have several protocol
primitives realized by different signaling message types. Each of
these message types may require a different amount of processing
power from the router.
Feher, Cselenyi, Korn Expires May 2002 [Page 10]
INTERNET-DRAFT <draft-ietf-bmwg-benchres-term-01.txt> November 2001
Measurement unit:
The signaling load is characterized by the signaling intensity,
which expresses how many signaling messages arrive to the router
within a time unit. The typical unit of the signaling intensity is
[1/s], which is the number of signaling messages that arrive
within one second.
6.3.4 Signaling Burst
Definition:
The signaling burst denotes a certain number of signaling messages
that arrive to the input port(s) of the router without
interruption, causing persistent load on the signaling message
handler.
Discussion:
Back-to-back signaling messages on one port of the router form a
typical signaling burst. However, other cases are imaginable, for
example when signaling messages arrive on different ports
simultaneously or with an overlap in time (i.e. when the tail of
one signaling message is behind the head of another one arriving
on another port).
Measurement unit:
The signaling burst is characterized by its length, which is the
number of messages that have arrived during the burst.
6.4 Performance Metrics
This group of definitions is the collection of the measurable impacts
that a resource reservation protocol has over the tested router
device it is running on.
6.4.1 Signaling Message Handling Time
Definition:
The signaling message handling time (or, in short, signal handling
time) is the time that a signaling message spends inside the
router before it is forwarded to the next node on the signaling
path.
Discussion:
Depending on the type of the signaling message, the router also
interprets the signaling messages, acts on them and forwards an
extended signaling message, which might contain information about
the result of the message processing. Thus the message handling
time is usually longer than forwarding time of data packets of the
same size. In addition, there might be also signaling message
primitives that are drained or generated by the router. Thus, the
signal handling time is defined as the time difference between the
time when a signaling message is received and the time the
corresponding processed signaling message is transmitted. If a
Feher, Cselenyi, Korn Expires May 2002 [Page 11]
INTERNET-DRAFT <draft-ietf-bmwg-benchres-term-01.txt> November 2001
message is not forwarded on the router, the signal handling time
is immeasurable; therefore it is not defined for such messages.
In the case of signaling messages that carry information
pertaining to multicast flows, the router might issue multiple
signaling messages after processing. In this case, by definition,
the signal handling time is the time interval elapsed between the
arrival of the incoming signaling message and the departure of the
last signaling message related to the received one.
This metric depends on the load on the router, as other tasks may
limit the processing power available to signaling message
handling. In addition to the router load, the signal handling time
may also be dependent on the type of the signaling message. For
example, it usually takes a shorter time for the router to tear
down a resource reservation session than to set it up.
Issues:
In the case of soft-state protocols, where refresh messages are
exchanged between peering network nodes only (see Reservation
Session Maintenance) the incoming refresh messages are drained by
the router making impossible to measure the signaling message
handling time on them.
Signal handling time is an important characteristic as it directly
affects the setup time of a session.
Measurement unit:
The typical unit of the signaling message handling time is
microsecond.
6.4.2 Premium Traffic Delay
Definition:
Premium traffic delay is the forwarding time of a packet that
belongs to a premium traffic flow passing through a resource
reservation capable router.
Discussion:
Premium traffic packets must be classified first in order to find
the resources dedicated to the flow. The time of the
classification is added to the usual forwarding time that a router
would spend on the packet without any resource reservation
capability.
There are routers where the processing power is shared between the
control plane and the data plane. This means that the processing
of signaling messages may have an impact on the data forwarding
performance of the router. In this case the premium traffic delay
metric reflects the influence the two planes have on each other.
Measurement unit:
The typical unit of the premium traffic delay is the microsecond.
Feher, Cselenyi, Korn Expires May 2002 [Page 12]
INTERNET-DRAFT <draft-ietf-bmwg-benchres-term-01.txt> November 2001
6.4.3 Best-effort Traffic Delay
Definition:
Best-effort traffic delay is the forwarding time of a packet that
does not belong to any premium traffic flow passing through a
resource reservation capable router.
Discussion:
It looks trivial that the classification algorithms do not have
any influence on the best-effort traffic. However, the processing
power sharing between the control and data plane may cause delays
in the forwarding procedure of each packet.
Measurement unit:
The typical unit of the best-effort traffic delay is microsecond.
6.4.4 Signaling Message Loss
Definition:
Signaling message loss is the ratio of the actual and the expected
number of signaling messages leaving a resource reservation
capable router subtracted from one.
Discussion:
There are certain types of signaling messages, which are required
to be forwarded by the router immediately when their processing is
finished. However, due to the high router load or for other
reasons, the forwarding or even the processing of the signaling
message might be canceled. To characterize such situations we
introduce the signaling message loss metric, which expresses the
ratio of the signaling messages that actually have left the router
and those ones that were expected to leave the router as a result
of the incoming sequence of signaling messages.
Since the most frequent reason for the signaling message loss is
the high router load, therefore this metric is suitable for
sounding out the scalability limits of resource reservation
capable routers.
Issues:
In the case of routers where network packets are queued in several
places, we have to be aware of that a signaling message may be
delayed seriously. Therefore, it may be hard or impossible to
determine whether the signaling message is still in the queues or
whether it was already dropped. By definition we say that a
signaling message is lost if there is no appearing forwarded
signaling message within a reasonable long time period. This time
period should be adjusted to the actual resource reservation
protocol (e.g. soft-state protocols may wait as much as the
refresh period to determine the loss of a signaling message).
Measurement unit:
Feher, Cselenyi, Korn Expires May 2002 [Page 13]
INTERNET-DRAFT <draft-ietf-bmwg-benchres-term-01.txt> November 2001
Usually, we measure the signaling message loss over a longer
period of time and then we express it as a percentage value.
Besides, in many cases it is enough to know that there was
signaling loss.
6.4.5 Session Refreshing Capacity
Definition:
The session refreshing capacity is the ratio of the truly
refreshed sessions and the number of session that have to be
refreshed during one refresh period. This metric is applied for
soft-state routers only.
Discussion:
The session refreshing capacity informs about condition of the
session maintenance. When the router is overloaded it may happen
that the router is not capable to refresh all the allocated
reservation sessions due to other tasks with higher priorities. In
this case sooner or later the resource reservation sessions over
the session refresh capacity are dropped even if the resource
reservation end-points are still refreshing them.
The session refreshing capacity sounds out the limit of resource
reservation session number that the router is capable to maintain.
Measurement unit:
The session refreshing capacity is expressed as a percentage
value.
6.4.6 Scalability Limit
Definition:
The scalability limit is the threshold between the steady state
and the overloaded state of the device under test.
Discussion:
All existing routers have finite buffer memory and finite
processing power. In the steady state of the router, the buffer
memories are not fully utilized and the processing power is enough
to cope with all tasks running on the router. As the router load
increases the router has to postpone more and more tasks. These
tasks (e.g. forwarding certain packets) are queued in the buffers,
and processed later. However, there is a certain point where no
more buffer memory is available; thus, the router becomes
overloaded and it is unable to store any more tasks for future
processing, so it is forced to drop them. Therefore the overloaded
state of the router can be recognized by the fact that some kind
of data (signaling or packet) loss occurs. A resource reservation
capable router may drop signaling messages, data packets or entire
resource reservation sessions.
The critical load condition when the router is still in the steady
state but the smallest amount of constant load increase would
Feher, Cselenyi, Korn Expires May 2002 [Page 14]
INTERNET-DRAFT <draft-ietf-bmwg-benchres-term-01.txt> November 2001
drive it to the overloaded state is the scalability limit of the
router.
7. Acknowledgement
We would like to thank the following individuals for their help in
forming this document: Joakim Bergkvist and Norbert Vegh from Telia
Research AB, Sweden, Krisztian Nemeth, Balazs Szabo, Gabor Kovacs and
Peter Vary from High Speed Networks Laboratory of Budapest University
of Technology and Economics, Hungary.
8. References
[1] Y. Bernet, et. al., "A Framework for Integrated Services
Operation over Diffserv Networks", RFC 2998, November 2000
[2] B. Braden, Ed., et. al., "Resource Reservation Protocol (RSVP) -
Version 1 Functional Specification", RFC 2205, September 1997.
[3] S. Bradner, "Benchmarking Terminology for Network
Interconnection Devices", RFC 1242, July 1991
[4] R. Mandeville, "Benchmarking Terminology for LAN Switching
Devices", RFC 2285, February 1998
[5] G. Feher, K. Nemeth, M. Maliosz, "Boomerang : A Simple Protocol
for Resource Reservation in IP Networks," Vancouver, IEEE Real-
Time Technology and Applications Symposium, June 1999
[6] P. Pan, H. Schulzrinne, "YESSIR: A Simple Reservation Mechanism
for the Internet", Computer Communication Review, on-line
version, volume 29, number 2, April 1999
[7] L. Delgrossi, L. Berger, "Internet Stream Protocol Version 2
(ST2) Protocol Specification - Version ST2+", RFC 1819, August
1995
[8] P. White, J. Crowcroft, "A Case for Dynamic Sender-Initiated
Reservation in the Internet", Journal on High Speed Networks,
Special Issue on QoS Routing and Signaling, Vol. 7 No. 2, 1998
[9] A. Eriksson, C. Gehrmann, "Robust and Secure Light-weight
Resource Reservation for Unicast IP Traffic", International WS
on QoS'98, IWQoS'98, May 18-20, 1998
[10] J. Wroclawski, "The Use of RSVP with IETF Integrated Services",
RFC 2210, September 1997
[11] K. Nichols et al., " Definition of the Differentiated Services
Field (DS Field) in the IPv4 and IPv6 Headers ", RFC 2474
9. Authors' Addresses:
Feher, Cselenyi, Korn Expires May 2002 [Page 15]
INTERNET-DRAFT <draft-ietf-bmwg-benchres-term-01.txt> November 2001
Gabor Feher
Budapest University of Technology and Economics (BUTE)
Department of Telecommunications and Telematics
Pazmany Peter Setany 1/D, H-1117, Budapest, Hungary
Phone: +36 1 463-3110
Email: feher@ttt-atm.ttt.bme.hu
Istvan Cselenyi
Telia Research AB
Vitsandsgatan 9B
SE 12386, Farsta
SWEDEN,
Phone: +46 8 713-8173
Email: istvan.i.cselenyi@telia.se
Andras Korn
Budapest University of Technology and Economics (BUTE)
Institute of Mathematics, Department of Analysis
Egry Jozsef u. 2, H-1111 Budapest, Hungary
Phone: +36 1 463-2475
Email: korn@math.bme.hu
Feher, Cselenyi, Korn Expires May 2002 [Page 16]