Network Working Group                                             S. Das
Internet-Draft                                              V. Narayanan
Intended status: Standards Track                              L. Dondeti
Expires: April 26, 2009                                   Qualcomm, Inc.
                                                        October 23, 2008

            ALTO: A Multi Dimensional Peer Selection Problem

Status of this Memo

   By submitting this Internet-Draft, each author represents that any
   applicable patent or other IPR claims of which he or she is aware
   have been or will be disclosed, and any of which he or she becomes
   aware will be disclosed, in accordance with Section 6 of 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-

   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

   The list of Internet-Draft Shadow Directories can be accessed at

   This Internet-Draft will expire on April 26, 2009.


   Bulk data applications are posing serious challenges to the Internet
   infrastructure and more mass adoption of such applications would only
   increase that challenge.  P2P bulk data applications and other large
   volume HTTP traffic such as video currently dominate the Internet
   traffic.  These applications do not generally benefit from the
   traffic engineering techniques developed for the Internet.  In the
   HTTP traffic case, the traffic optimization issues are often
   addressed using the CDN infrastructure.  For P2P bulk data
   applications, the problem is about picking the right peers to
   communicate with and the criteria for doing this vary greatly based
   on the application.  Hence, intelligent peer selection is the

Das, et al.              Expires April 26, 2009                 [Page 1]

Internet-Draft                    multi                     October 2008

   fundamental problem to address for these applications.  This document
   explains the multiple dimensions of the peer selection problem with
   respect to obtaining information from the network as well from other
   peers and argues for an integrated, common framework to share such

Table of Contents

   1.  Introduction . . . . . . . . . . . . . . . . . . . . . . . . .  3
   2.  Terminology  . . . . . . . . . . . . . . . . . . . . . . . . .  5
   3.  Problem Statement  . . . . . . . . . . . . . . . . . . . . . .  5
     3.1.  Network-assisted ALTO  . . . . . . . . . . . . . . . . . .  6
     3.2.  Peer-assisted ALTO . . . . . . . . . . . . . . . . . . . .  6
     3.3.  The need for multiple criterion  . . . . . . . . . . . . .  6
       3.3.1.  Peer-assisted ALTO alone . . . . . . . . . . . . . . .  7
       3.3.2.  Network-assisted ALTO alone  . . . . . . . . . . . . .  7
     3.4.  Applicability and Discussion . . . . . . . . . . . . . . .  8
   4.  Design Goals . . . . . . . . . . . . . . . . . . . . . . . . .  9
   5.  Security Considerations  . . . . . . . . . . . . . . . . . . . 11
   6.  IANA Considerations  . . . . . . . . . . . . . . . . . . . . . 11
   7.  Acknowledgments  . . . . . . . . . . . . . . . . . . . . . . . 11
   8.  References . . . . . . . . . . . . . . . . . . . . . . . . . . 12
     8.1.  Normative References . . . . . . . . . . . . . . . . . . . 12
     8.2.  Informative References . . . . . . . . . . . . . . . . . . 12
   Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 13
   Intellectual Property and Copyright Statements . . . . . . . . . . 14

Das, et al.              Expires April 26, 2009                 [Page 2]

Internet-Draft                    multi                     October 2008

1.  Introduction

   A large portion of the Internet traffic today is from P2P (peer-to-
   peer) applications.  Such an architecture for data transfer is
   attractive because it reduces the bandwidth costs of the content
   provider since they simply need to seed the content to a few nodes in
   the network which would then contribute upload bandwidth to assist
   the content provider's servers in the data transfer.  Thus, the
   single point bottleneck at the content provider is eliminated and
   both users and content providers benefit.

   However such an approach is detrimental to another important entity
   in the system: the ISPs.  While p2p has led to increased popularity
   of broadband connections and arguably increased subscribers for ISPs;
   it has also increased traffic costs for the ISPs.  This is because
   P2P applications' peer selection does not consider underlying network
   topology and link costs in that topology.  Most p2p applications
   typically are only interested in improving their data transfer
   performance which is satisfied if the download link of the user is
   saturated.  As shown in [2], traffic generated by popular P2P
   applications often cross network boundaries multiple times,
   overloading links which are frequently subject to congestion [3].
   This happens because most p2p applications simply use random peer
   selection followed by monitoring and reevaluation.  Even if p2p
   applications perform some network measurements, these typically tend
   to be round trip time estimation which may or may not lead to peer
   selection conducive to ISP interests.

   This led to ISPs efforts at shaping or blocking P2P traffic on
   specific ports.  In response, p2p applictions started using dynamic
   ports to transfer data following whcih ISPs had to use deep packet
   protocol specific information to shape p2p flows.  In response, p2p
   applications have started encrypting their connections.  The use of
   TCP RST messages to deter costly p2p application data transfer has
   led to some conflicts as well.  It is in the ISP's interests to avoid
   the cat-and-mouse games of protocol-specific detection and mitigation
   while still not having to increase costs significantly to accomodate
   p2p traffic.

   One way to reduce the impact on ISPs would be the deployment of
   caching entities in the networks to reduce cross-ISP traffic and
   network distance of data transfer.  However such an approach has
   several issues:

   o  It is not clear who would deploy these high-bandwidth large-
      storage capable caches since it can be argued that caching pushes
      the costs from the content provider to the ISPs

Das, et al.              Expires April 26, 2009                 [Page 3]

Internet-Draft                    multi                     October 2008

   o  The caches would need to be able to cache data from all p2p
      applications and consequently become complicated to deploy and
      maintain over time as p2p application evolve

   o  The use of caches opens up the issue of storage of copyrighted

   Thus, a solution is needed that can allow for peer selection by the
   p2p application themselves that is friendly to the ISP's network
   costs while being friendly to the applications' objective of good
   performance for the user.

   Recent studies [4], [5], [6] have suggested that it may be possible
   to reduce the impact that P2P applications have on ISPs traffic
   costs.  This is mainly achieved by informed peer selection in the P2P
   applications guided by network level metrics instead of random
   selection.  However, p2p applications do not have a trust
   relationship with network operators and what may be good for the ISP
   is not necessarily good for the performance of the p2p applications.
   These competing interests necessitate a solution for ALTO that
   addresses the interests of both the entities in the system.

   This document describes the problem of optimizing traffic generated
   by P2P applications using information provided by third parties, i.e.
   the other peers in the network or the network operator.  The overall
   goal of optimizing the P2P application is for them to become more
   network-friendly, while at the same time allowing the networks to
   remain application-friendly.  The following are key arguments that we
   put forth in this draft:

   o  The problem of peer selection needs to take into account the
      interests of the ISPs, application providers and the peers.  The
      goal should be to reduce the impact on ISP without sacrificing the
      performance of p2p applications.  There are many situations we
      elaborate upon in Section 3 where simply considering ISP interests
      leads to poor performance for p2p applications.  Information about
      peer connectivity characteristics is an important component of the
      ALTO problem of peer selection (in conjunction with ISP routing
      information) and together with network topology information, can
      enable peers to optimize their performance as well as take into
      account ISP interests.

   o  An ALTO system with information about ISP routing alone does not
      provide enough incentives for adoption in p2p applications, due to
      the competing interests of the two parties and the lack of trust.
      Combining both elements of peer selection creates an incentive for
      p2p applications to actually use the ALTO service.

Das, et al.              Expires April 26, 2009                 [Page 4]

Internet-Draft                    multi                     October 2008

   o  Application-specific solutions for sharing peer connectivity
      characteristics are inefficient and cause unnecessary overhead in
      the network.  A common information plane allows reuse of such
      information across a wide range of applications ranging from DHT
      next hop selection, multi-source file download, relay selection,
      mirror selection, etc., since many of the connectivity
      characteristics of peers such as upload/download bandwidth,
      network coordinate, wireless link information are useful to a
      multitude of p2p applications in peer selection.

   The rest of this draft is structured as follows.  Section 3
   introduces the problem and argues for the multiple criterion that
   need to be considered when developing solutions, while Section 4
   describes the design goals of the system.

2.  Terminology

   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
   document are to be interpreted as described in RFC 2119 [1].

3.  Problem Statement

   Bulk data applications are posing serious challenges to the Internet
   infrastructure and more mass adoption of such applications would only
   increase that challenge.  P2P bulk data applications and other large
   volume HTTP traffic such as video currently dominate the Internet
   traffic.  These applications do not generally benefit from the
   traffic engineering techniques developed for the Internet.  In the
   HTTP traffic case, the traffic optimization issues are often
   addressed using the CDN infrastructure.  For P2P bulk data
   applications, the problem is about picking the right peers to
   communicate with and the criteria for doing this vary greatly based
   on the application.  Hence, intelligent peer selection is the
   fundamental problem to address for these applications.

   One necessary input for intelligent peer selection has been shown to
   be network topology information.  And, accurate network topology
   information can only be obtained with the help of ISPs.  However,
   there is more to peer selection than just network topology

   Consider a Client Application at node A which wants to access a
   service for peer selection such as ALTO.  Overall peer selection in
   ALTO can be performed using two sets of information: (1) One set of
   information is that from the network, i.e. network-assisted ALTO

Das, et al.              Expires April 26, 2009                 [Page 5]

Internet-Draft                    multi                     October 2008

   which promotes peer selection that is beneficial to the ISP by either
   reducing the interdomain traffic or reducing congestion on bottleneck
   links inside the ISP. (2) The other set of information is from the
   set of potential Client Application Providers are targets for peer
   connectivity, i.e. peer-assisted ALTO.

3.1.  Network-assisted ALTO

   Network-assisted application layer traffic optimization refers to the
   use of network operators in guiding peer selection for p2p
   applications.  Network operators can use network topology and traffic
   statistics to provide hints to the P2P applications on which hosts it
   would prefer the application pick for data transfer.  For example,
   peers that involve inter-domain routing would be given lower priority
   in the response to the querying P2P application.  Another example
   would be that peers whose connectivity is such that their choice
   would increase the congestion on an already congested link inside the
   ISP.  Such peers would also be given lower priority.  Some recent
   work has proposed solutions in this space [5],[4].

3.2.  Peer-assisted ALTO

   Peer-assisted application layer traffic optimization refers to the
   use of published information about peers and their connectivity to
   the Internet in guiding peer selection for p2p applications.  For
   example, BitTorrent-like data transfer applications would query
   metrics related to the peer of interest to decide on the peers to
   select for initial data transfer.  This information could be for
   example the upload bandwidth of the peer.  These choices could then
   be refined or changed by monitoring data connections.  Another
   example would be for VoIP applications to query the ALTO service for
   peers that it wants to use as relays to find one with low latency.
   Altough latency information is pairwise the peer could publish its
   network coordinate calculated by a system such as GNP (Global Network
   Positioning) into the ALTO service which could then be retrieved and
   used by a querying peer to estimate network latency.  Publishing peer
   connectivity could also involve edge-specific information such as
   which cell id a peer is connected to in a cellular network so that
   another peer in the same cell is discouraged from making a connection
   within the same cell to minimize the congestion and avoid occuping 2
   downlink and 2 uplink channels in the same cell.  This type of
   traffic optimization cannot be acheived via network-assisted ALTO.

3.3.  The need for multiple criterion

   This section justifies the need for multiple criterion in ALTO: i.e.
   the need for both ISP related information and peer related
   information.  Note that the underlying assumption is that peers and

Das, et al.              Expires April 26, 2009                 [Page 6]

Internet-Draft                    multi                     October 2008

   ISPs may not necessarily trust each other and thus any solution in
   this space needs to consider the interests of both parties.

3.3.1.  Peer-assisted ALTO alone

   It is already well known that using peer related information alone is
   not enough.  Simply randomly selecting peers and trying them out and
   reevaluating the choices based on observed performance is suboptimal
   as it takes time to converge on a good set of peers as well as causes
   unnecessary overhead in the network.  Incorporating locality into
   peer selection using either active measurements or some sort of
   information plane (e.g. [7]) has been shown to improve the download
   completion time performance of BitTorrent.  For example, [7] showed a
   35% reduction in download time for 60% of the peers in a swarm.
   Other studies [5] showed this gain to be around 10-25%.  However, [5]
   also showed that such an approach increases the traffic on congestion
   on bottleneck links by 69%.  In summary, while peer related
   information is powerful in improving performance, it affects ISP
   performance and can lead to shaping and throttling.

3.3.2.  Network-assisted ALTO alone

   On the other hand there are several situations where relying on ISP
   information by itself is not sufficient.  If ISPs simply return a
   rank ordered list of node identifiers in response to a list of node
   identifiers sent in a query, the querying node may not be optimizing
   its performance.  It is possible that the peers in a swarm that are
   within the ISP are all DSL hosts which can upload at 128Kbps while
   some of the other peers that may involve interdomain routing are
   actually high bandwidth fiber or cable connected hosts with high
   upload bandwidth.  In such cases, the p2p application user would
   sustain high download times to meet the ISP's objective.  Similarly,
   peers that are nearby geographically (which is correlated with
   latency typically) may be preferred for relays in VoIP but do not
   meet the ISP's objective since they use a different access network.

   The reasons for these problems are two-fold: (1) The ISPs' objectives
   in general may not always match the users objective of improved
   performance, and (2) The ISP can only provide an indication of its
   preference for connectivity to certain peers outside its domain but
   it does not have any notion of end-to-end performance provided by
   these outside peers.

   Thus neither network-asissted ALTO or peer-assisted ALTO alone help
   in bringing together these competing interests together.  Thus, in
   this draft we argue for a combined solution that provides a true
   information plane for peer selection, i.e. one that does not
   sacrifice but, in fact improves the performance of p2p applications

Das, et al.              Expires April 26, 2009                 [Page 7]

Internet-Draft                    multi                     October 2008

   while getting cooperation from p2p applications for the ISP.  ISP
   information is only one of the two pillars of the ALTO service - it
   is an important one, given that the ISPs have knowledge of internal
   congestion, interdomain peering policies and connectivity
   information.  Peer information is the other pillar of the ALTO
   service; and by providing it as a service to all p2p applications we
   reduce the overhead and inefficiency of each application performing
   their own measurements which is inefficient for the network and
   complicated for application developers.

3.4.  Applicability and Discussion

   The problem of peer selection has been studied at various dimensions
   and has pretty broad applicability.  At the very basic level,
   intelligent peer selection can be applied, in file sharing networks,
   to the problem of choosing the right source to download a file from.
   However, there are a variety of other applications for such
   intelligent peer selection.  For example, locality aware structured
   overlays are built with topology-assisted neighbor selection models
   or topology-based identity generation shcemes.  The more accurate the
   topology information is, the more efficient such schemes will be.
   Very fundamentally, intelligent neighbor selection schemes aim at
   improving the query latency in such systems.  Peer selection is also
   applicable to other problems such as identifying the best super peer
   to contact in a system, using the best relay for NAT traversal,
   identifying the best next hop for a query based on several criteria
   (e.g., quality, reputation, semantic expertise, etc.), etc.

   A lot of study has been done on the applicability of the peer
   selection problem.  Some examples include [8], [9], [10], [11], [12],

   Two observations can be made from these applicability aspects of peer
   selection.  One is that the problem is applicable to a wide range of
   scenarios with some possible generalization - in all cases, while the
   criteria for peer selection and the context in which it is done may
   be different, the basic model for peer selection revolves around
   ranking a list of peers based on information collected about the
   peers in accordance with the given criteria.  A second observation is
   that the same piece of information collected could be put to use in
   very different ways in different scenarios - e.g., the use of
   topology information in determining source peers to download content
   from and neighbor peers to make connections with are very different

   Peer selection is also equally important for application providers,
   ISPs and the peers themselves for very different reasons.  ISPs have
   an incentive to keep their operational costs to a minimum, while

Das, et al.              Expires April 26, 2009                 [Page 8]

Internet-Draft                    multi                     October 2008

   ensuring their customers a good experience.  Application providers
   are interested in keeping their own operational costs to a minimum,
   which by nature, conflicts with the ISP's interests of keeping the
   costs low.  Application providers are also interested in ensuring
   that the application performs well to keep the peers interested in
   the application.  For the peers themselves, the emphasis is on
   performance as well as their own access costs.

   The varied scenarios of applicability of peer selection information,
   the diverse interests of various parties in peer selection and the
   underlying common nature of these models suggest that the problem
   should be tackled in a uniform manner, allowing for generality and
   extensibility of information.  The scenarios also broadly fall under
   two buckets - one, where information may be obtained from specific
   hosts (e.g., ISP hosts, reputation tracking servers, etc.) and
   another, where information may be obtained from any number of peers.
   The final peer selection algorithm may be a combination of various
   pieces of such information in a manner that fits a given scenario and
   criterion.  Solving any one piece of this puzzle separately is likely
   to lead to multiple different protocols and mechanisms for the same
   problem.  Further, any one such protocol is unlikely to result in a
   dramatic improvement of the scenario (be it bandwidth utilization or
   latency reduction, etc.).  For instance, as argued earlier in this
   draft, network-assistance and peer information are both important to
   consider for peer selection.  Hence, the adoption of any one of these
   protocols will be less incentivized in such a fragmented model.

   The ALTO service, given a set of host location identifiers for peers,
   may annotate them with ISP information (such as the p-distance from
   P4P) as well as peer connectivity information (which is variable and
   could be related to bandwidth, network coordinates, wireless link
   information etc ) and return the annotated set to the p2p
   application.  The p2p application can use this data to perform peer
   selection based on its requirements.  We argue that the peer
   connectivity information be a generic container that can contain
   different types of information.  However the request response
   protocol can be unified as an ALTO protocol for retrieving the
   annotated information for each destination peer.  Note that the data
   pertaining peer-assisted ALTO and network-assisted ALTO may be stored
   differently but the interface to p2p applications can be kept

4.  Design Goals

   The fundamental goal of ALTO, extracting from what has been described
   thus far, has to do with allowing the use of both network-assisted
   and peer-assisted information exchanges for various categories of

Das, et al.              Expires April 26, 2009                 [Page 9]

Internet-Draft                    multi                     October 2008

   applications that can use it.  The main idea is to model ALTO as a
   service that these applications may use to apply the peer selection
   intelligence in the specific context of interest to the application.
   The following are design goals for the ALTO service framework:

   1.  The ALTO service should accommodate specific hosts providing
       network topology information.

          Network topology information may be provided by hosts in an
          ISP network with much greater accuracy than can be done via
          other schemes.  Hence, a model that allows for such
          information exchange from specific hosts is useful.  The
          discovery of such hosts capable of providing this information
          falls in this territory as well.

   2.  The ALTO service should allow peers to publish ALTO related
       information in an application agnostic manner.

          Peers may publish information of various kinds that helps in
          peer selection.  A primary example of such information is the
          connectivity characteristics of a peer.  This type of
          information allows other criteria to be taken into account for
          peer selection, such as the uplink/downlink bandwidth of
          peers, the wireless nature of links, etc.  This information,
          coupled with topology or proximity information, will allow
          exchange of bulk data in a manner that benefits the ISPs and
          the peers.

   3.  The ALTO service should allow for both centralized and
       distributed service models.

          ISP hosts and any servers that are queried for ALTO related
          information may be offering the ALTO service in a centralized
          model.  On the other hand, peer information related to ALTO
          may be obtained in a distributed fashion, e.g., by querying an
          overlay.  It is important to cater to both models and also
          allow for the two models to be linked.  For instance, a query
          on an overlay for ALTO related information may result in a
          different node on the overlay querying the ISP's ALTO host.

   4.  The ALTO protocols should be agnostic to the actual peer
       selection algorithm in use.

          The peer selection algorithm itself is contextual and depends
          on different factors for different scenarios.  For instance,
          source selection for bulk data downloads involves optimal
          bandwidth usage and performance as criteria, neighbor
          selection in structured overlays may be a function of hop

Das, et al.              Expires April 26, 2009                [Page 10]

Internet-Draft                    multi                     October 2008

          count and delay, relay selection for NAT traversal may be a
          function of the relay capacity and connectivity
          characteristics and so on.  Hence, the ALTO protocols should
          be agnostic to the specific peer selection algorithm for any
          given instantiation.

   5.  The ALTO protocols should be generic to allow any type of
       information useful for performing ALTO services to be exchanged.
       The ALTO protocols should be extensible to allow carrying new
       types of information that may be defined at a future point.

          As in the case of the algorithm, the types of information
          exchanged for peer selection are also contextual to various
          scenarios.  For instance, bandwidth or latency based peer
          selection may involve topology information and connectivity
          characteristics of peers, relevance based peer selection may
          involve the semantic expertise of various peers, and so on.
          Hence, the ALTO protocols should simply be a container to
          transport information that assists in peer selection without
          being dependent on the actual type of information exchanged.

5.  Security Considerations

   Information exchange, as facilitated by ALTO, may be sensitive and
   may require secure communication.  Hence, the ALTO protocols must
   provide for channel security properties such as confidentiality and
   integrity protection.  Further, the exchange of some of this
   information may need to be limited to certain entities.  This calls
   for authentication and authorization of entities prior to exchange of
   ALTO information.  Hence, the ALTO framework should provide for
   entity authentication and authorization as part of the service.  When
   various types of information to be carried in the ALTO protocols are
   standardized, a call must be made about whether it is subject to the
   authorization model.  Further, some types of information exchanges
   could lead to privacy issues to various parties and the implications
   of that need to be studied prior to standardization.

6.  IANA Considerations

   This document does not have any IANA considerations.

7.  Acknowledgments

   This draft has benefited from the insights presented in several IETF
   drafts namely the ALTO problem statement, requirements and survey

Das, et al.              Expires April 26, 2009                [Page 11]

Internet-Draft                    multi                     October 2008

   drafts.  The research work done on IPlane, P4P, Taming the torrent,
   and ISP and P2P oracles as well as the various measurement studies on
   P2P applications impact on ISPs performed by the research community
   have helped inform us in creating this draft.

8.  References

8.1.  Normative References

   [1]   Bradner, S., "Key words for use in RFCs to Indicate Requirement
         Levels", BCP 14, RFC 2119, March 1997.

8.2.  Informative References

   [2]   Karagiannis, T., Rodriguez, P., and K. Papagiannaki, "Should
         ISPs fear Peer-Assisted Content Distribution?", IMC Proceedings
         of the Internet Measurement Conference, 2005.

   [3]   Akella, A., Seshan, S., and A. Shaikh, "An Empirical Evaluation
         of WideArea Internet Bottlenecks",  Proceedings of ACM SIGCOMM,

   [4]   Aggarwal, V., Feldmann, A., and C. Scheideler, "Can ISPs and
         P2P systems co-operate for improved performance?",  ACM SIGCOMM
         Computer Communications Review (CCR), 37:3, pp. 29-40, 2006.

   [5]   Xie, H., Krishnamurthy, A., Silberschatz, A., and R. Yang,
         "P4P: Explicit Communications for Cooperative Control Between
         P2P and Network Providers",  Proceedings of ACM SIGCOMM, 2008.

   [6]   Choffnes, D. and F. Bustamante, "Taming the Torrent: A
         practical approach to reducing cross-ISP traffic in P2P
         systems",  Proceedings of ACM SIGCOMM, 2008.

   [7]   Madhyastha, H., Isdal, T., Piatek, M., Dixon, C., Anderson, T.,
         Krishnamurthy, A., and A. Venkataramani, "iPlane: an
         information plane for distributed services",  Proceedings of
         USENIX OSDI, 2006.

   [8]   Lo, V., Liu, Y., GauthierDickey, C., and J. Li, "Scalable
         Leader Selection in Peer-to-Peer Overlay Networks",
          Proceedings of Second International Workshop on Hot Topics in
         Peer-to-Peer Systems (Hot-P2P), 2005.

   [9]   Xhafa, F., Barolli, L., Fernandez, R., and T. Daradoumis, "An
         Experimental Study on Peer Selection in a P2P Network over
         PlanetLab",  Proceedings of International Conference on

Das, et al.              Expires April 26, 2009                [Page 12]

Internet-Draft                    multi                     October 2008

         Parallel Processing Workshops, 2007.

   [10]  Yang, X. and G. Veciana, "Service Capacity of Peer to Peer
         Networks",  Proceedings of IEEE INFOCOM, 2004.

   [11]  Habib, A. and J. Chuang, "Service Differentiated Peer
         Selection: An Incentive Mechanism for Peer-to-Peer Media
         Streaming",  Proceedings of IWQoS, 2004.

   [12]  Tang, S., Wang, H., and P. Meigham, "The Effect of Peer
         Selection with Hopcount or Delay Constraint on Peer-to-Peer
         Networking",  Proceedings of IFIP NETWORKING, 2008.

   [13]  Haas, P. and R. Siebes, "Peer Selection in PeertoPeer Networks
         with Semantic Topologies",  Proceedings of WWW, 2004.

Authors' Addresses

   Saumitra M. Das
   Qualcomm, Inc.
   3195 Kifer Road
   Santa Clara, CA

   Phone: +1 408-533-9529

   Vidya Narayanan
   Qualcomm, Inc.
   5775 Morehouse Dr
   San Diego, CA

   Phone: +1 858-845-2483

   Lakshminath Dondeti
   Qualcomm, Inc.
   5775 Morehouse Dr
   San Diego, CA

   Phone: +1 858-845-1267

Das, et al.              Expires April 26, 2009                [Page 13]

Internet-Draft                    multi                     October 2008

Full Copyright Statement

   Copyright (C) The IETF Trust (2008).

   This document is subject to the rights, licenses and restrictions
   contained in BCP 78, and except as set forth therein, the authors
   retain all their rights.

   This document and the information contained herein are provided on an

Intellectual Property

   The IETF takes no position regarding the validity or scope of any
   Intellectual Property Rights or other rights that might be claimed to
   pertain to the implementation or use of the technology described in
   this document or the extent to which any license under such rights
   might or might not be available; nor does it represent that it has
   made any independent effort to identify any such rights.  Information
   on the procedures with respect to rights in RFC documents can be
   found in BCP 78 and BCP 79.

   Copies of IPR disclosures made to the IETF Secretariat and any
   assurances of licenses to be made available, or the result of an
   attempt made to obtain a general license or permission for the use of
   such proprietary rights by implementers or users of this
   specification can be obtained from the IETF on-line IPR repository at

   The IETF invites any interested party to bring to its attention any
   copyrights, patents or patent applications, or other proprietary
   rights that may cover technology that may be required to implement
   this standard.  Please address the information to the IETF at

Das, et al.              Expires April 26, 2009                [Page 14]