Internet DRAFT - draft-jacquenet-mpls-rd-p2mp-te-requirements

draft-jacquenet-mpls-rd-p2mp-te-requirements







Internet Engineering Task Force                             C. Jacquenet
Internet-Draft                                     France Telecom Orange
Intended status: Informational                                   Q. Zhao
Expires: January 11, 2014                              Huawei Technology
                                                                B. Zhang
                                                    Telus Communications
                                                                 E. Metz
                                                                     KPN
                                                           July 10, 2013


Requirements for Receiver-Driven Traffic Engineered Point-to-Multi-Point
                      (P2MP) MPLS Tree Structures
          draft-jacquenet-mpls-rd-p2mp-te-requirements-03.txt

Abstract

   This document presents a set of requirements for the establishment
   and maintenance of Receiver-Driven Point-to-Multipoint (RD-P2MP) and
   Multipoint-to-Multipoint (RD-MP2MP) Traffic-Engineered (TE)
   Multiprotocol Label Switching (MPLS) Label Switched Paths (LSPs).

Status of This Memo

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

   Internet-Drafts are working documents of the Internet Engineering
   Task Force (IETF).  Note that other groups may also distribute
   working documents as Internet-Drafts.  The list of current Internet-
   Drafts is at http://datatracker.ietf.org/drafts/current/.

   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."

   This Internet-Draft will expire on January 11, 2014.

Copyright Notice

   Copyright (c) 2013 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
   (http://trustee.ietf.org/license-info) in effect on the date of
   publication of this document.  Please review these documents



Jacquenet, et al.       Expires January 11, 2014                [Page 1]

Internet-Draft    Receiver-Driven P2MP LSP Requirements        July 2013


   carefully, as they describe your rights and restrictions with respect
   to this document.  Code Components extracted from this document must
   include Simplified BSD License text as described in Section 4.e of
   the Trust Legal Provisions and are provided without warranty as
   described in the Simplified BSD License.

Table of Contents

   1.  Introduction  . . . . . . . . . . . . . . . . . . . . . . . .   2
     1.1.  Rationale . . . . . . . . . . . . . . . . . . . . . . . .   2
     1.2.  Typical Use Case: Interconnecting PIM Islands Through A
           MPLS Backbone . . . . . . . . . . . . . . . . . . . . . .   3
     1.3.  Scope and Assumptions . . . . . . . . . . . . . . . . . .   4
   2.  Terminology . . . . . . . . . . . . . . . . . . . . . . . . .   4
   3.  Requirements  . . . . . . . . . . . . . . . . . . . . . . . .   5
     3.1.  Operation . . . . . . . . . . . . . . . . . . . . . . . .   5
     3.2.  Forwarding  . . . . . . . . . . . . . . . . . . . . . . .   6
     3.3.  Signaling . . . . . . . . . . . . . . . . . . . . . . . .   7
     3.4.  Robustness  . . . . . . . . . . . . . . . . . . . . . . .   7
     3.5.  Performance and Scalability . . . . . . . . . . . . . . .   8
     3.6.  Management  . . . . . . . . . . . . . . . . . . . . . . .  10
     3.7.  Backward Compatibility  . . . . . . . . . . . . . . . . .  10
   4.  Acknowledgements  . . . . . . . . . . . . . . . . . . . . . .  10
   5.  IANA Considerations . . . . . . . . . . . . . . . . . . . . .  11
   6.  Security Considerations . . . . . . . . . . . . . . . . . . .  11
   7.  References  . . . . . . . . . . . . . . . . . . . . . . . . .  11
     7.1.  Normative References  . . . . . . . . . . . . . . . . . .  12
     7.2.  Informative References  . . . . . . . . . . . . . . . . .  13
   Authors' Addresses  . . . . . . . . . . . . . . . . . . . . . . .  13

1.  Introduction

1.1.  Rationale

   The dramatic growth of multicast-based IP services such as live TV
   broadcasting raises new challenges for network operators.  The sole
   use of IP multicast becomes challenging, given the QoS-demanding
   nature of the applications, especially in light of the deployment of
   High Definition or 3D TV.

   The specification of traffic-engineered, MPLS-based, Point-to-Multi-
   Point (P2MP) tree structures [6] is meant to address both quality and
   robustness issues, but failed to be massively adopted and deployed by
   operators so far, mostly because the current standard assumes a
   priori knowledge of all the routers involved in the establishment and
   the maintenance of the tree structure, at the cost of extra
   management complexity.




Jacquenet, et al.       Expires January 11, 2014                [Page 2]

Internet-Draft    Receiver-Driven P2MP LSP Requirements        July 2013


   Current technological state-of-the-art shows that most of
   implementations assume the static configuration of the basic
   information that will be used by routers to build a traffic-
   engineered P2MP LSP path a la [6].  Scalability issues may also be
   raised by the current standard as core routers are likely to maintain
   a number of RSVP states that will grow with the number of leaf
   routers.

   The receiver-initiated paradigm of IP multicast is also a true
   incentive for faster convergence in MPLS environments, yielding an
   improved computation and establishment of traffic-engineered P2MP
   MPLS tree structures.

   In addition, the inability to take into account the network access
   capabilities of the receivers, so that the computation of the tree
   structure can dynamically adapt to such access conditions becomes
   even more challenging with the ever-growing number of multicast
   channels that need to be delivered to the receivers.

   Thus, we believe that the combination of the dynamics of receiver-
   initiated IP multicast distribution trees with the hard QoS
   guarantees brought by a MPLS-based traffic engineered tree
   computation scheme is promising, for the sake of optimized
   convergence and facilitated (if not automated) hard-guaranteed
   service design and operation.

   Taking the best of breed between PIM-computed [16], receiver-driven
   multicast distribution trees and MPLS traffic engineering
   capabilities can indeed significantly improve

1.2.  Typical Use Case: Interconnecting PIM Islands Through A MPLS
      Backbone

   Taking the best of breed between PIM-computed, receiver-driven
   multicast distribution trees and MPLS traffic engineering
   capabilities can indeed significantly improve the level of quality
   associated to the delivery of multicast services.

   One typical example would be a network design where PIM "islands"
   (for example, PIM-enabled access infratsructures that connect
   multicast receivers, as well as the IPTV head end that connects the
   multicast source) connect to a MPLS backbone infrastructure.
   Multicast traffic is forwarded along the RD-P2MP LSP paths without
   soliciting an additional routing protocol such as BGP ([14], [15]) to
   learn the multicast routes and exchange the PIM multicast states of
   each PIM island, so that P2MP LSP paths can be established
   accordingly.




Jacquenet, et al.       Expires January 11, 2014                [Page 3]

Internet-Draft    Receiver-Driven P2MP LSP Requirements        July 2013


   The Receiver-Driven approach should solely rely upon the PIM protocol
   and the use of a PIM/MPLS inter-working function at the border
   between the PIM islands and the MPLS backbone infrastructure, so that
   no PIM state has to be maintained by the MPLS routers of the
   backbone.

1.3.  Scope and Assumptions

   This document details the requirements for the dynamic establishment
   and maintenance of receiver-driven, traffic-engineered Point-to-
   Multi-Point (RD-P2MP) tree structures.  As such, considerations about
   Label Distribution Protocol (LDP) extensions for the computation of
   P2MP tree structures [10] are out of the scope of this draft.

   Likewise, this document exclusively focuses on traffic-engineered
   P2MP MPLS tree structures.  As such, traffic-engineered Multi-Point-
   to-Multi-Point (MP2MP) MPLS tree structures are not taken into
   consideration, mostly because it is believed that the primary
   applicability of MPLS traffic engineering capabilities for multicast-
   enabled services remains IPTV services that assume a 1:N group
   communication scheme that is typical of Source-Specific Multicast
   designs.  Nevertheless, the requirements detailed in this document
   are also valid for receiver-driven, traffic-engineered MP2MP tree
   structures.

2.  Terminology

   The following terms are used in this document:

   o  Sender: The source of the content, as in [9].

   o  Receiver: The Receiver of the content multicast by the source, as
      in [9].

   o  Upstream: The direction of a given traffic from a Receiver towards
      a Sender, as defined in [9].

   o  Downstream: The direction of multicast traffic from a Sender
      towards a Receiver, as defined in [9].

   o  Path-Sender: The sender of a RSVP_PATH message, with NO
      correlation to the directionality of the corresponding multicast
      traffic.

   o  Path-Receiver: The receiver of a RSVP_PATH message, with NO
      correlation to the directionality of the corresponding multicast
      traffic.




Jacquenet, et al.       Expires January 11, 2014                [Page 4]

Internet-Draft    Receiver-Driven P2MP LSP Requirements        July 2013


   o  Path-Initiator: The Path-Sender that originated a RSVP_PATH
      message.  A Path-Initiator is different from a Path-Sender in that
      an intermediate node (a node that is part of the receiver-driven
      P2MP tree structure) can be a Path-Sender.

   o  Path-Terminator: The Path-Receiver that does NOT propagate a
      RSVP_PATH message.  A Path-Terminator is different from the Path-
      Receiver in that an intermediate node (a node that is part of the
      receiver-driven P2MP tree structure) can be a Path-Receiver.

   o  Receiver-Driven P2MP Label Switched Path (RD P2MP LSP): A traffic-
      engineered P2MP MPLS Label Switched Path that is dynamically
      computed and established at the initiative of the receivers.

3.  Requirements

   [5] specifies requirements for P2MP traffic-engineered MPLS Label
   Switched Paths.  These requirements are equally applicable to
   Receiver-Driven traffic-engineered MPLS Label Switched Paths.  This
   section details the additional requirements raised by the dynamic
   computation and establishment of RD P2MP LSPs.

3.1.  Operation

   REQ-1  Grafting and pruning of branches of any given RD P2MP tree
      structure MUST be dynamic and receiver-initiated.

   REQ-2  Leaves of RD P2MP LSP paths MUST be aware of the corresponding
      P2MP LSP identifiers for tree computation and maintenance
      purposes.

   REQ-3  A leaf router MUST initiate RSVP_PATH messages that will be
      sent towards the root router.  RSVP_PATH messages are triggered by
      typical signaling subcription procedures originated by receivers,
      such as IGMP/MLD Report messages that may be directly processed by
      the leaf routers of a given RD P2MP LSP.

   REQ-4  A node that receives a RSVP_PATH message MUST first decide if
      this message will make itself a branch Label Switch Router (LSR)
      or not.  In the case that it will become a transit LSR because of
      this RSVP_PATH message, the router will allocate required
      resources on the interface through which the RSVP_PATH message is
      received, before forwarding it upstream (for the sake of multicast
      traffic delivery efficiency).

   REQ-5  In case the node that received the RSVP_PATH message is
      already a branch or transit node for the multicast content
      associated to the said RSVP_PATH message, the node MUST allocate



Jacquenet, et al.       Expires January 11, 2014                [Page 5]

Internet-Draft    Receiver-Driven P2MP LSP Requirements        July 2013


      required resources (if available) on the interface through which
      the RSVP_PATH message is received.  This node MUST NOT send the
      RSVP_PATH message upstream.

   REQ-6  Upon receipt of a RSVP_PATH message, a Path-Terminator MUST
      send back a RSVP_RESV message that MUST be forwarded along the RD
      P2MP LSP to the Path-Sender.

   REQ-7  A node receiving a RSVP_RESV message SHOULD interpret it as a
      successful resource reservation from the upstream node for the
      establishment of the RD P2MP LSP.

   REQ-8  The termination of a L2S (Leaf-to-Source) sub-path that
      belongs to a given RD P2MP LSP MUST be one of the ingress routers
      where multicast data sent by a source enter the said RD P2MP LSP.

   REQ-9  Label allocation MUST be done prior to sending RSVP_PATH
      messages upstream.

   REQ-10  To facilitate the computation of RD P2MP LSP paths within a
      network whose LSRs do not all support the same capabilities with
      respect to RD P2MP LSP signaling and data forwarding, the
      capability of a given LSR to support the RSVP-TE signaling and
      forwarding features for RD P2MP LSP path computation purposes MUST
      be advertised to its neighbor LSRs.

3.2.  Forwarding

   REQ-11  Just like typical P2P MPLS [3], any given multicast traffic
      characterized by a multicast group address is associated with a
      FEC (Forwarding Equivalence Class).  All packets that belong to a
      particular FEC MUST be forwarded along the corresponding RD P2MP
      LSP.

   REQ-12  RD P2MPLSPs MAY be deployed over multi-access media such as
      Ethernet.  In these environments, it is possible that the entry
      point to the network segment is a branch LSR of the RD P2MPLSP.
      To avoid all replicated data are sent through the same port and
      carried on the same segment, a mechanism SHOULD be supported by
      the said branch LSR so as to send a single copy of the multicast
      data onto the multi-access network to reach several downstream
      nodes

   REQ-13  The RD P2MP LSP computation scheme SHOULD provide a means for
      a Branch LSR to send a single copy of the multicast data through
      an Ethernet LAN interface to reach several downstream nodes.





Jacquenet, et al.       Expires January 11, 2014                [Page 6]

Internet-Draft    Receiver-Driven P2MP LSP Requirements        July 2013


   REQ-14  As a consequence of the previous requirement, the same label
      MUST be negotiated with all downstream LSRs for any given RD P2MP
      LSP.

   REQ-15  When there are several candidate upstream LSRs conected to a
      given LAN segment, the RD P2MP LSP path computation scheme SHOULD
      provide a means for all downstream LSRs to select the same
      upstream LSR, so as to avoid traffic replication.

   REQ-16  The RD P2MP LSP path computation scheme SHOULD allow for an
      efficient balancing of a set of P2MP LSPs among a set of candidate
      upstream LSRs connected to a LAN segment.

3.3.  Signaling

   Certain parameters (such as priority and bandwidth) are associated
   with an LSP.  The parameters are installed by means of the RSVP
   signalling specific to the establishment and the maintenance of RD
   P2MP LSPS.  As such:

   REQ-17  Downstream or upstream LSRs MUST NOT alter the attributes set
      and signaled by a leaf router of any given RD P2MP LSP.

   REQ-18  A consistent QoS policy SHOULD be enforced from the root to
      all leaves of a single P2MP/MP2MP LSP.

   REQ-19  Some leaves of a given tree may yield the enforcement of a
      different QoS policy, depending on the various access capabilities
      of the receivers.  Still, content will be delivered to these
      receivers by using the same (core) P2MP/MP2MP tree structure.

   REQ-20  Changing the parameters for the whole tree MAY be supported,
      but the change MUST apply to the whole tree from ingress LSR to
      all egress LSRs.

3.4.  Robustness

   REQ-21  The detection of a more optimal path (e.g., from a cost
      standpoint) is an example of a situation where re-routing of the
      RD P2MP LSP MAY be required.  While re-routing is in progress,
      duplicate bandwidth reservation over the common parts between the
      old and new RD P2MP LSP MUST be avoided.

   REQ-22  RD P2MP LSP path computation, establishment and maintenance
      MUST support a Make-Before-Break procedure to ensure that there is
      minimal traffic disruption during re-routing operations.





Jacquenet, et al.       Expires January 11, 2014                [Page 7]

Internet-Draft    Receiver-Driven P2MP LSP Requirements        July 2013


   REQ-23  Scope of Make-Before-Break procedures MUST be restricted to
      the relevant sub-LSP portion of any given RD P2MP LSP, for the
      sake of resource optimization and overall service performance.

   REQ-24  The support of a Make-Before-Break procedure MUST include re-
      optimization facilities for any impacted sub-LSP portion of a
      given RD P2MP LSP.

   REQ-25  Any Make-Before-Break operation MUST not impact the rest of
      the RD P2MP LSP, from both a signaling and operational
      standpoints.

   REQ-26  Where sub-LSP re-optimization is allowed by the ingress LSR,
      such re- optimization MAY be initiated by a downstream LSR that is
      the root of the sub-LSP to be re-optimized.  Sub-LSP re-
      optimization initiated by a downstream LSR MUST be carried out
      without dramatically impacting the forwarding of multicast traffic
      along the corresponding RD P2MP LSP.

   REQ-27  Any downstream node located on the sub-LSP that is being re-
      optimized SHOULD have control on the re-optimization procedure.

   REQ-28  Overtime, a given optimized path may become sub-optimized.
      Path re-computation capabilities SHOULD be supported and they
      could be triggered by updated IGP cost metrics, time interval or a
      combination thereof.

   REQ-29  Traffic load balancing capabilities SHOULD be supported by
      the RD P2MP LSP path computation scheme, to enforce least-fill-
      based load sharing computation strategies.

3.5.  Performance and Scalability

   REQ-30  The RD P2MP LSP computation, establishment and maintenance
      MUST be scalable: The switching performances of the routers that
      are part of a RD P2MP LSP MUST NOT be affected during RD P2MP LSP
      path computation, establishment and operation.

   REQ-31  Likewise, scalability of RD P2MP LSP path design and
      operation MUST NOT be jeopardized by the number of receivers, nor
      by the number of RD P2MP LSPs maintained at any given time by the
      routers of the network. in particular, RD P2MP LSP path
      computation, establishment and operation SHOULD accommodate the
      need to deliver multicast traffic to several millions of
      receivers, without any perceived overall service performance
      degradtion, including the preservation of customer's Quality of
      Experience.




Jacquenet, et al.       Expires January 11, 2014                [Page 8]

Internet-Draft    Receiver-Driven P2MP LSP Requirements        July 2013


   REQ-32  Dynamic grafting and pruning operations pertaining to any
      given RD P2MP LSP MUST NOT affect multicast traffic forwarding
      efficiency, from a packet loss and one-way transit delay
      perspectives, among other possible Quality of Service metrics.

   REQ-33  Dynamic grafting and pruning operations pertaining to any
      given RD P2MP LSP SHOULD NOT infer any other additional processing
      than the processing specific to the portion of the RD P2MP LSP
      from the added/removed leaf LSR to the corresponding branch LSR.

   REQ-34  Dynamic grafting and pruning operations pertaining to any
      given RD P2MP LSP MUST NOT disrupt the forwarding of multicast
      traffic along the RD P2MP LSP at any given time.

   REQ-35  In order to scale to a large number of branches, RD P2MP LSPs
      SHOULD be unambiguously identified by means of P2MP ID identifiers
      that MUST be persistent for the whole RD P2MP LSP, regardless of
      the number of branches and leaves.

   REQ-36  In order to accommodate a growing number of leaves, the
      amount of a P2MP LSP states on a given LSR, for one particular RD
      P2MPLSP SHOULD only depend on the number of adjacent LSRs that
      belong to the same RD P2MP LSP.

   REQ-37  RD P2MP LSP performance and scalability assessment SHOULD
      rely upon the following metrics (and a combination thereof):

   o  Number of receivers

   o  Number of sources

   o  Number of leaf routers

   o  Number of branch routers

   o  Number of branches

   o  Number of RD P2MP LSPs to be deployed over the MPLS network

   REQ-38  Scalability of control plane operation (setup, maintenance,
      modification, and teardown) MUST be considered.  In particular:

   o  The amount of refresh processing associated with maintaining a RD
      P2MP LSP.

   o  The amount of protocol state that must be maintained by ingress
      and transit LSRs along a RD P2MP LSP.




Jacquenet, et al.       Expires January 11, 2014                [Page 9]

Internet-Draft    Receiver-Driven P2MP LSP Requirements        July 2013


   o  The number of protocol messages required to set up or tear down a
      RD P2MP LSP as a function of the number of egress LSRs.

   o  The number of protocol messages required to repair a RD P2MP LSP
      after failure or to perform make-before-break.

   o  The amount of protocol information transmitted to manage a RD P2MP
      LSP (i.e., the amount of management traffic as a function of the
      global bandwidth resources available).

   o  The amount of signaling traffic required for RD P2MP LSP path
      compuation, setup and operation.

   o  The amount of additional control plane processing required in the
      network to detect whether an add/delete of a new branch is
      required, and in particular, the amount of processing in steady
      state when no add/delete is requested.

   o  The amount of control plane processing required by the ingress,
      transit, and egress LSRs to add/delete a branch LSP to/from an
      existing RD P2MP LSP.

3.6.  Management

   REQ-39  Design and operation of RD P2MP TE LSP paths MUST support
      management capabilities as per the Specific Management Functional
      Areas (SMFAs), namely Fault, Configuration, Accounting,
      Performance and Security management capabilities.  In particular,
      RD P2MP TE LSP-based and Source-to-Leaf (S2L) traffic statistics
      management MUST be supported.

3.7.  Backward Compatibility

   REQ-40  The RD P2MP LSP path computation scheme MUST allow the
      continued use of existing techniques to establish P2P and legacy
      P2MP LSPs (Traffic-engineered or not) within the same network,.and
      MUST allow the coexistence of Receiver-Driven P2MP/MP2MP LSPSs
      with P2P and legacy P2MP LSPs within the same network.

   REQ-41  RD P2MP LSP paths MUST be able to coexist with legacy P2P and
      P2MP LSPs within the same network.

   REQ-42  Signaling capabilities of the RD P2MP LSP path computation
      scheme MUST NOT prevent the signaling of "legacy" P2P and P2MP LSP
      paths.

4.  Acknowledgements




Jacquenet, et al.       Expires January 11, 2014               [Page 10]

Internet-Draft    Receiver-Driven P2MP LSP Requirements        July 2013


   We would like to thank authors of [5] and [13] who inspired some of
   the text of this draft.

5.  IANA Considerations

   This draft makes no request to IANA.

6.  Security Considerations

   This document does not define any protocol extensions and does not,
   therefore, make any changes to any security models.  It is a
   requirement that any RD P2MP LSP design MUST include mechanisms to
   enable the secure establishment and management. of RD P2MP LSPs.
   This means in particular that:

   o  A receiver MUST be authenticated before it is allowed to trigger
      the grafting of an additional leaf of a RD P2MP LSP tree
      structure,

   o  Mechanisms to provide some guarantees about the identity of an
      ingress LSR that belongs to a given RD P2MP LSP SHOULD be
      supported,

   o  Mechanisms to ensure that communicating signaling entities can
      verify each other's identities SHOULD be supported,

   o  Mechanisms to ensure that control plane messages are protected
      against spoofing and tampering SHOULD be supported,

   o  Mechanisms to ensure that unauthorized leaves or branches are not
      added to any given RD P2MP/ LSP; and mechanisms to protect
      signaling messages from snooping MUST be supported.

   Note that RD P2MP LSP signaling mechanisms built on P2P RSVP-TE
   signaling and RSVP-TE P2MP signalling are likely to inherit all the
   security techniques and problems associated with RSVP-TE.  These
   problems may be exacerbated in P2MP situations where security
   associations may need to be maintained between any given ingress LSR
   and multiple egress LSRs.  Such issues are similar to security issues
   raised by the IP multicast transmission scheme.

7.  References









Jacquenet, et al.       Expires January 11, 2014               [Page 11]

Internet-Draft    Receiver-Driven P2MP LSP Requirements        July 2013


7.1.  Normative References

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

   [2]        Awduche, D., Malcolm, J., Agogbua, J., O'Dell, M., and J.
              McManus, "Requirements for Traffic Engineering Over MPLS",
              RFC 2702, September 1999.

   [3]        Rosen, E., Viswanathan, A., and R. Callon, "Multiprotocol
              Label Switching Architecture", RFC 3031, January 2001.

   [4]        Awduche, D., Berger, L., Gan, D., Li, T., Srinivasan, V.,
              and G. Swallow, "RSVP-TE: Extensions to RSVP for LSP
              Tunnels", RFC 3209, December 2001.

   [5]        Yasukawa, S., "Signaling Requirements for Point-to-
              Multipoint Traffic-Engineered MPLS Label Switched Paths
              (LSPs)", RFC 4461, April 2006.

   [6]        Aggarwal, R., Papadimitriou, D., and S. Yasukawa,
              "Extensions to Resource Reservation Protocol - Traffic
              Engineering (RSVP-TE) for Point-to-Multipoint TE Label
              Switched Paths (LSPs)", RFC 4875, May 2007.

   [7]        Farrel, A., Papadimitriou, D., Vasseur, J., and A.
              Ayyangar, "Encoding of Attributes for Multiprotocol Label
              Switching (MPLS) Label Switched Path (LSP) Establishment
              Using Resource ReserVation Protocol-Traffic Engineering
              (RSVP-TE)", RFC 4420, February 2006.

   [8]        Kompella, K. and Y. Rekhter, "Label Switched Paths (LSP)
              Hierarchy with Generalized Multi-Protocol Label Switching
              (GMPLS) Traffic Engineering (TE)", RFC 4206, October 2005.

   [9]        Braden, B., Zhang, L., Berson, S., Herzog, S., and S.
              Jamin, "Resource ReSerVation Protocol (RSVP) -- Version 1
              Functional Specification", RFC 2205, September 1997.

   [10]       Wijnands, IJ., Minei, I., Kompella, K., and B. Thomas,
              "Label Distribution Protocol Extensions for Point-to-
              Multipoint and Multipoint-to-Multipoint Label Switched
              Paths", RFC 6388, November 2011.

   [11]       Pan, P., Swallow, G., and A. Atlas, "Fast Reroute
              Extensions to RSVP-TE for LSP Tunnels", RFC 4090, May
              2005.




Jacquenet, et al.       Expires January 11, 2014               [Page 12]

Internet-Draft    Receiver-Driven P2MP LSP Requirements        July 2013


   [12]       Berger, L., "Generalized Multi-Protocol Label Switching
              (GMPLS) Signaling Functional Description", RFC 3471,
              January 2003.

   [13]       Le Roux, JL. and T. Morin, "Requirements for Point-to-
              Multipoint Extensions to the Label Distribution Protocol",
              RFC 6348, September 2011.

   [14]       Rosen, E. and R. Aggarwal, "Multicast in MPLS/BGP IP
              VPNs", RFC 6513, February 2012.

   [15]       Aggarwal, R., Rosen, E., Morin, T., and Y. Rekhter, "BGP
              Encodings and Procedures for Multicast in MPLS/BGP IP
              VPNs", RFC 6514, February 2012.

   [16]       Fenner, B., Handley, M., Holbrook, H., and I. Kouvelas,
              "Protocol Independent Multicast - Sparse Mode (PIM-SM):
              Protocol Specification (Revised)", RFC 4601, August 2006.

7.2.  Informative References

   [17]       Andersson, L. and G. Swallow, "The Multiprotocol Label
              Switching (MPLS) Working Group decision on MPLS signaling
              protocols", RFC 3468, February 2003.

   [18]       Berger, L., "Generalized Multi-Protocol Label Switching
              (GMPLS) Signaling Resource ReserVation Protocol-Traffic
              Engineering (RSVP-TE) Extensions", RFC 3473, January 2003.

   [19]       Le Faucheur, F. and W. Lai, "Requirements for Support of
              Differentiated Services-aware MPLS Traffic Engineering",
              RFC 3564, July 2003.

   [20]       Berger, L., Takacs, A., Caviglia, D., Fedyk, D., and J.
              Meuric, "GMPLS Asymmetric Bandwidth Bidirectional Label
              Switched Paths (LSPs)", RFC 5467, March 2009.

Authors' Addresses

   Christian Jacquenet
   France Telecom Orange
   4 rue du Clos Courtel
   35512 Cesson Sevigne
   France

   Phone: +33 2 99 12 43 82
   Email: christian.jacquenet@orange.com




Jacquenet, et al.       Expires January 11, 2014               [Page 13]

Internet-Draft    Receiver-Driven P2MP LSP Requirements        July 2013


   Quintin Zhao
   Huawei Technology
   125 Nagog Park
   Acton, MA  01919
   US

   Email: qzhao@huawei.com


   Boris Zhang
   Telus Communications
   200 Consilium Pl. Floor 15
   Toronto, ON  M1H 3J3
   Canada

   Email: Boris.Zhang@telus.com


   Eduard Metz
   KPN
   Netherlands

   Email: eduard.metz@kpn.com




























Jacquenet, et al.       Expires January 11, 2014               [Page 14]