Seamoby Working Group                                        H. Jung
Internet Draft                                               S.  Koh
Document:draft-jung-seamoby-paging-buffload-00.txt        ETRI/KOREA
Category: Informational                                 December 2001


        Reducing the Buffering Load of Paging Agent in IP Paging
              <draft-jung-seamoby-paging-buffload-00.txt>

Status of this Memo

   This document is an Internet-Draft and is in full conformance with
   all provisions of Section 10 of RFC 2026.

   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

Convention used in this draft

   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 RFC 2119.

Abstract

   This draft defines two procedures to distribute buffering load in IP
   paging using PA.  A couple of documents using the PA, which includes
   all functions defined RFC 3154 in an entity, have beee submitted for
   cadidate protocol for IP paging. In this PA paging approach, the
   centralized buffering in PA can be a bottleneck point if a PA has to
   manage large paging area and a number of idle mobile hosts.  To
   address the problem, it can be good solution to distribute buffering
   load among ARs in the paging area.


1. Introduction

   IP paging has been designed to decrease the signaling load for
   location registration and to prevent rapid power consumption of
   mobile nodes.  For these purposes, the IETF SeaMoby WG has recently
   produced substantial documents, including Problem Statement and
   Requirements for IP paging protocol [1, 2].  Based on the functional



H. Jung, et al. draft-jung-seamoby-paging-buffload-00.txt       [Page 1]


INTERNET-DRAFT             Expires: June 2002              December 2001


   requirements, specific proposals for IP paging protocol have been
   made in the Working Group.  In some IP paging schemes currently
   considered, an agent called Paging Agent(PA), or Paging Function(PF),
   is employed to perform all the functionalities required for IP paging
   which are specified in RFC3154 [3, 4].

   In this document, we consider the buffering functionality of PA,
   which is one of the IP paging functionalities.  The PA temporarily
   buffers the data destined to a idle MN, until the location of MN is
   known to PA.  In this approach, a PA must perform the buffering for
   all of the MNs in the network.  Therefore, the PA is likely to be
   overloaded due to the buffering operations for all the MNs in the
   network.  The buffering load at the PA becomes more severe, as the
   size of paging area gets larger and the number of idle MNs in the
   area increases.  In this document, we propose an alternate buffering
   scheme designed to ensure that the buffering load of PA can be
   reduced significantly, compared to the existing scheme.  The main
   idea of the proposed scheme is that the buffering for a MN is
   performed by its last-registered AR other than PA.  The last-
   registered AR represents an AR which a MN has registered to, after it
   entered the current paging area.  By this, the buffering load of PA
   can be distributed into the boundary ARs of the paging area.


2.  Buffering Scheme Using the Last-Registered ARs

   The proposed scheme is designed based on the principle that the
   processing load of PA can be alleviated by distributing its buffering
   load over the ARs(or Foreign Agents) in the paging area, which can
   also be described in the literature [5, 6].  In the proposed scheme,
   the buffering functionality is performed by ARs other than PA,
   diffferently from the existing paging schemes.  Figure 1 illustrates
   the distribution of buffering load using the last-registered ARs in
   the paging area.  We assume that MN's idle registration is
   acomplished by registrating its serving PA address as CoA to HA.
   Then the location update will be done only if the serving PA is
   changed.

   In the figure a MN registered its idle state to PA through AR-A.  It
   means that AR-A is the last-registered AR of the MN.  We assume that
   the MN is now located in the coverage of AR-D by way of AR-B and AR-
   C.  When the PA receives data packets destined to idle MN from
   HA(1a), it realizes that the MN is in the idle state, and it thus
   forwards the recevied packets to the last registered AR, AR-A(1b).
   The AR-A will save these packets in its buffer.  The PA now initiates
   the paging procedure(1c) to locate the MN in the paging area.

   Each router in the area will receive the L3 paging message and then



H. Jung, et al. draft-jung-seamoby-paging-buffload-00.txt       [Page 2]


INTERNET-DRAFT             Expires: June 2002              December 2001


   translate it into the corresponding L2 paging signaling through air
   interface(2).

   The MN receives the paging signal and then responds with the
   registration message to the last-registered AR-A(3b) as well as the
   PA(3a), through its currently serving AR-D.  The last-registered AR-A
   will then forward the buffered data packets to the MN by way of AR-
   D(4).  Hereafter, the MN comes back to the active state, and thus it
   will receive the subsequent data packets directly from HA, not via
   the last-registered AR-A.


                                |
                                | (1a)
                                V
                             +------+
                             |      |
                             |  PA  |<--------------+
                             |      |               |
                             +------+               |
                               | |                  |
                           (1b)| |                  |
                  +------------+ |                  |
                  |              |(1c)              |
                  |              |                  |(3a)
                  |              |                  |
             +----|--------------|------------------|--------+
             |    | +---------+--+------+-----------|-+      |
             |    | |         |  (3b)   |           | |      |
             |    | | +---------------------------+-+ |      |
             |    | | |       |         |      (4)| | |      |
             |    V V V       V         V         V | V      |
             |  +------+   +------+   +------+   +------+    |
             |  | AR-A |   | AR-B |   | AR-C |   | AR-D |    |
             |  +------+   +------+   +------+   +------+    |
             |      |(2)       |(2)       |(2)     ^ |(2)    |
             |      |          |          |        | | ^     |
             |      V          V          V        | V |     |
             |                                  +------+     |
             |                                  |  MN  |     |
             |                                  +------+     |
             |                                               |
             | IP paging area                                |
             +-----------------------------------------------+

             Figure 1: Buffering Using the Last-Registered AR





H. Jung, et al. draft-jung-seamoby-paging-buffload-00.txt       [Page 3]


INTERNET-DRAFT             Expires: June 2002              December 2001


   The only requirement to implement the proposed scheme is that each of
   ARs in the area needs to have the buffer capacity.  On the other
   hand, it is relatively easy to extend the paging signaling algorithm
   and messages to support the proposed scheme.

3. Further Enhancement of the Proposed Scheme

   The proposed scheme is effective in that the buffering load of PA can
   be alleviated.  However, it is noted that the border routers of the
   paging area are usually the last-registered ARs, and thus the
   buffering loads for the area may be concentrated on these border ARs.
   Another concern of the scheme is that a failure of the
   buffering(last-registered) AR, which may not be on the data path,
   induces overall failure of data delivery.

   One possible way to address the problems described above is to select
   and use the most suitable AR in the paging area as the buffering AR
   for each MN.  The selection of the most suitable AR will be made by
   PA, based on the state information gathered by interaction between PA
   and candidate ARs in the area.  Such state information may be the
   simple status of AR(running or failed), or may include more detailed
   system status such as available buffer memory and current processing
   loads.

   In any way, the PA needs to collect the state information from ARs in
   the area.  On the reverse, each AR must report its state information
   to the PA.  For this purpose, a new signaling mechanism may be
   specified along with signaling messages on the basis of periodic time
   or a pre-configured rule.  Alternatively, an existing IP routing or
   management protocols such as RIP, OSPF, SNMP may be used or extended
   for conveying the status information.

   In the enhanced scheme using the most suitable AR, the buffering
   procedures are similar to those using the last-registered AR, other
   than the following differences:

   When PA receives the packet destined to MN from HA, it first selects
   the best candidate AR as the buffering AR for the MN, based on the
   collected state information.

   After that, the PA begins to locate the idle MN by sending paging
   signal message.  The paging message must contain information(like an
   identifier) to indicate which AR is serving for the MN as the
   buffering AR in the area.

   In resonse to the paging message, the MN sends registration message
   to PA as well as its buffering AR.  The buffering AR will then
   forwards the buffered packets to MN.



H. Jung, et al. draft-jung-seamoby-paging-buffload-00.txt       [Page 4]


INTERNET-DRAFT             Expires: June 2002              December 2001


4. Implementation Notes

   TBD


5. Security Considerations

   TBD


6.  References

   [1] J. Kempf.  Dormant mode host alerting(ip paging) problem
   statement.  Request for Comments 3132, Internet Engineering Task
   Force, June 2001.

   [2] J. Kempf, C. Castelluccia, et. al., Requirements and Functional
   Architecture for an IP Host Alerting Protocol, Request for Comments
   3154, Internet Engineering Task Force, August 2001

   [3] Rajeev Koodli, et. al., Dormant Mode Handover Support in Mobile
   Networks, Internet draft, draft-koodli-paging-01.txt, Internet
   Engineering Task Force, May 2002

   [4] G. Renker, et. al., Paging Concept for IP based Networks,
   internet draft, draft-renker-paging-ipv6-00.txt, Internet Engineering
   Task Force, June 2001

   [5] R. Ramjee, et. al., IP Paging Service for Mobile Hosts, ACM
   MOBICOM, 2001

   [6] X. Zhang, et. al., P-MIP: Paging in Mobile IP, The Fourth
   International Workshop on Wireless Mobile Multimedia, Rome, Italy,
   July 2001

















H. Jung, et al. draft-jung-seamoby-paging-buffload-00.txt       [Page 5]


INTERNET-DRAFT             Expires: June 2002              December 2001


7.  Authors' Addresses

   HeeYoung Jung
   Protocol Engineering Center
   Electronics and Telecommunication Research Institute
   161 Kajung-Dong, Yusong-Gu, Taejon, Korea
   Phone:  +82 42 860-4928
   EMail:  hyjung@etri.re.kr
   Fax:    +82 42 861-5404

   Seok Joo Koh
   Protocol Engineering Center
   Electronics and Telecommunication Research Institute
   161 Kajung-Dong, Yusong-Gu, Taejon, Korea
   Phone:  +82 42 860-6218
   EMail:  sjkoh@etri.re.kr
   Fax:    +82 42 861-5404




Full Copyright Statement

   "Copyright(C) The Internet Society(2000).  All Rights Reserved.  This
   document and translations of it may be copied and furnished to others, and
   derivative works that comment on or otherwise explain it or assist in its
   implementation may be prepared, copied, published and distributed, in whole
   or in part, without restriction of any kind, provided that the above
   copyright notice and this paragraph are included on all such copies and
   derivative works.  However, this document itself may not be modified in any
   way, such as by removing the copyright notice or references to the Internet
   Society or other Internet organizations, except as needed for the purpose of
   developing Internet standards in which case the procedures for copyrights
   defined in the Internet Standards process must be followed, or as required
   to translate it into languages other than English.

   The limited permissions granted above are perpetual and will not be
   revoked by the Internet Society or its successors or assigns.

   This document and the information contained herein is provided on an "AS IS"
   basis and THE INTERNET SOCIETY AND THE INTERNET ENGINEERING TASK FORCE
   DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO
   ANY WARRANTY THAT THE USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY
   RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A
   PARTICULAR PURPOSE."






H. Jung, et al. draft-jung-seamoby-paging-buffload-00.txt       [Page 6]