RFC 4216






Network Working Group                                    R. Zhang, Ed.
Request for Comments: 4216                Infonet Services Corporation
Category: Informational                             J.-P. Vasseur, Ed.
                                                   Cisco Systems, Inc.
                                                         November 2005


                   MPLS Inter-Autonomous System (AS)
                 Traffic Engineering (TE) Requirements

Status of This Memo



   This memo provides information for the Internet community.  It does
   not specify an Internet standard of any kind.  Distribution of this
   memo is unlimited.

Copyright Notice



   Copyright (C) The Internet Society (2005).

Abstract



   This document discusses requirements for the support of inter-AS MPLS
   Traffic Engineering (MPLS TE).  Its main objective is to present a
   set of requirements and scenarios which would result in general
   guidelines for the definition, selection, and specification
   development for any technical solution(s) meeting these requirements
   and supporting the scenarios.

Table of Contents



   1. Introduction ....................................................3
      1.1. Conventions Used in This Document ..........................3
   2. Contributing Authors ............................................4
   3. Definitions and Requirements Statement ..........................5
      3.1. Definitions ................................................5
      3.2. Objectives and Requirements of Inter-AS Traffic
           Engineering ................................................7
           3.2.1. Inter-AS Bandwidth Guarantees .......................7
           3.2.2. Inter-AS Resource Optimization ......................8
           3.2.3. Fast Recovery across ASes ...........................8
      3.3. Inter-AS Traffic Engineering Requirements Statement ........9
   4. Application Scenarios ...........................................9
      4.1. Application Scenarios Requiring Inter-AS Bandwidth
           Guarantees .................................................9
           4.1.1. Scenario I - Extended or Virtual PoP (VPoP) .........9
           4.1.2. Scenario II - Extended or Virtual Trunk ............11




Zhang & Vasseur              Informational                      [Page 1]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


           4.1.3. Scenario III - End-to-End Inter-AS MPLS TE
                  from CE to CE ......................................12
      4.2. Application Scenarios Requiring Inter-AS Resource
           Optimization ..............................................13
           4.2.1. Scenario IV - TE across multi-AS within a
                  Single SP ..........................................13
           4.2.2. Scenario V - Transit ASes as Primary and
                  Redundant Transport ................................14
   5. Detailed Requirements for Inter-AS MPLS Traffic Engineering ....16
      5.1. Requirements within One SP Administrative Domain ..........16
           5.1.1. Inter-AS MPLS TE Operations and Interoperability ...16
           5.1.2. Protocol Signaling and Path Computations ...........16
           5.1.3. Optimality .........................................17
           5.1.4. Support of Diversely Routed Inter-AS TE LSP ........17
           5.1.5. Re-Optimization ....................................18
           5.1.6. Fast Recovery Support Using MPLS TE Fast Reroute ...18
           5.1.7. DS-TE Support ......................................18
           5.1.8. Scalability and Hierarchical LSP Support ...........19
           5.1.9. Mapping of Traffic onto Inter-AS MPLS TE Tunnels ...19
           5.1.10. Inter-AS MPLS TE Management .......................19
                  5.1.10.1. Inter-AS MPLS TE MIB Requirements ........19
                  5.1.10.2. Inter-AS MPLS TE Fault Management
                            Requirements .............................20
           5.1.11. Extensibility .....................................21
           5.1.12. Complexity and Risks ..............................21
           5.1.13. Backward Compatibility ............................21
           5.1.14. Performance .......................................21
      5.2. Requirements for Inter-AS MPLS TE across Multiple SP ......22
           5.2.1. Confidentiality ....................................22
           5.2.2. Policy Control .....................................23
                  5.2.2.1. Inter-AS TE Agreement Enforcement
                           Polices ...................................23
                  5.2.2.2. Inter-AS TE Rewrite Policies ..............24
                  5.2.2.3. Inter-AS Traffic Policing .................24
   6. Security Considerations ........................................24
   7. Acknowledgements ...............................................24
   8. Normative References ...........................................25
   9. Informative References .........................................25
   Appendix A. Brief Description of BGP-based Inter-AS Traffic
               Engineering ...........................................27











Zhang & Vasseur              Informational                      [Page 2]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


1.  Introduction



   The MPLS Traffic Engineering (TE) mechanism documented in [TE-RSVP]
   may be deployed by Service Providers (SPs) to achieve some of the
   most important objectives of network traffic engineering as described
   in [TE-OVW].  These objectives are summarized as:

   - Supporting end-to-end services requiring Quality of Service (QoS)
     guarantees
   - Performing network resource optimization
   - Providing fast recovery

   However, this traffic engineering mechanism can only be used within
   an Autonomous System (AS).

   This document discusses requirements for an inter-AS MPLS Traffic
   Engineering mechanism that may be used to achieve the same set of
   objectives across AS boundaries within or beyond an SP's
   administrative domains.

   The document will also present a set of application scenarios where
   the inter-AS traffic engineering mechanism may be required.  This
   mechanism could be implemented based upon the requirements presented
   in this document.

   These application scenarios will also facilitate discussions for a
   detailed requirements list for this inter-AS Traffic Engineering
   mechanism.

   Please note that there are other means of traffic engineering
   including Interior Gateway Protocol (IGP); metrics-based (for use
   within an AS); and Border Gateway Protocol (BGP) attribute-based (for
   use across ASes, as described in Appendix A), which provide coarser
   control of traffic paths.  However, this document addresses
   requirements for a MPLS-based, fine-grained approach for inter-AS TE.

   This document doesn't make any claims with respect to whether it is
   possible to have a practical solution that meets all the requirements
   listed in this document.

1.1.  Conventions Used in This Document



   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].






Zhang & Vasseur              Informational                      [Page 3]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


2.  Contributing Authors



   The co-authors listed below contributed to the text and content of
   this document.  (The contact information for the editors appears in
   section 9, and is not repeated below.)

   Kenji Kumaki
   KDDI Corporation
   Garden Air Tower
   Iidabashi, Chiyoda-ku,
   Tokyo 102-8460, JAPAN
   EMail : ke-kumaki@kddi.com

   Paul Mabey
   Qwest Communications
   950 17th Street,
   Denver, CO 80202, USA
   EMail: pmabey@qwest.com

   Nadim Constantine
   Infonet Services Corporation
   2160 E. Grand Ave.
   El Segundo, CA 90025. USA
   EMail: nadim_constantine@infonet.com

   Pierre Merckx
   EQUANT
   1041 route des Dolines - BP 347
   06906 SOPHIA ANTIPOLIS Cedex, FRANCE
   EMail: pierre.merckx@equant.com

   Ting Wo Chung
   Bell Canada
   181 Bay Street, Suite 350
   Toronto, Ontario, Canada, M5J 2T3
   EMail: ting_wo.chung@bell.ca

   Jean-Louis Le Roux
   France Telecom
   2, avenue Pierre-Marzin
   22307 Lannion Cedex, France
   EMail: jeanlouis.leroux@francetelecom.com

   Yonghwan Kim
   SBC Laboratories, Inc.
   4698 Willow Road
   Pleasanton, CA 94588, USA
   EMail: Yonghwan_Kim@labs.sbc.com



Zhang & Vasseur              Informational                      [Page 4]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


3.  Definitions and Requirements Statement



3.1.  Definitions



   The following provides a list of abbreviations and acronyms
   specifically pertaining to this document:

   SP:               Service Providers including regional or global
                     providers.

   SP Administrative
   Domain:           a single SP administration over a network or
                     networks that may consist of one AS or multiple
                     ASes.

   IP-only networks: SP's network where IP routing protocols such as
                     IGP/BGP are activated.

   IP/MPLS networks: SP's network where MPLS switching capabilities and
                     signaling controls (e.g., ones described in
                     [MPLS-ARCH]) are activated in addition to IP
                     routing protocols.

   Intra-AS TE:      A generic definition for traffic engineering
                     mechanisms operating over IP-only and/or IP/MPLS
                     network within an AS.

   Inter-AS TE:      A generic definition for traffic engineering
                     mechanisms operating over IP-only and/or IP/MPLS
                     network across one or multiple ASes.  Since this
                     document only addresses IP/MPLS networks, any
                     reference to Inter-AS TE in this document refers
                     only to IP/MPLS networks and is not intended to
                     address IP-only TE requirements.

   TE LSP:           MPLS Traffic Engineering Label Switched Path.

   Intra-AS MPLS TE: An MPLS Traffic Engineering mechanism where its TE
                     Label Switched Path (LSP), Head-end Label Switching
                     Router (LSR), and Tail-end LSR reside in the same
                     AS for traffic engineering purposes.

   Inter-AS MPLS TE: An MPLS Traffic Engineering mechanism where its TE
                     LSPs, Head-end LSR, and Tail-end LSR do not reside
                     within the same AS or both Head-end LSR and Tail-
                     end LSR are in the same AS, but the TE LSP
                     transiting path may be across different ASes.




Zhang & Vasseur              Informational                      [Page 5]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


   ASBRs:            Autonomous System Border Routers used to connect to
                     another AS of a different or the same Service
                     Provider via one or more links that interconnect
                     ASes.

   Inter-AS TE Path: A TE path traversing multiple ASes and ASBRs, e.g.,
                     AS1-ASBR1-inter-AS link(s)-ASBR2-AS2... ASBRn-ASn.

   Inter-AS TE
   Segment:          A portion of the Inter-AS TE path.

   Inter-AS DS-TE:   Diffserv-aware Inter-AS TE.

   CE:               Customer Edge Equipment

   PE:               Provider Edge Equipment that has direct connections
                     to CEs.

   P:                Provider Equipment that has backbone trunk
                     connections only.



   VRF:              Virtual Private Network (VPN) Routing and
                     Forwarding Instance.

   PoP:              Point of presence or a node in SP's network.

   SRLG:             A set of links may constitute a 'shared risk link
                     group' (SRLG) if they share a resource whose
                     failure may affect all links in the set as defined
                     in [GMPLS-ROUT].

   PCC:              Path Computation Client; any client application
                     requesting a path computation to be performed by
                     the Path Computation Element.

   PCE:              Path Computation Element; an entity (component,
                     application or network node) that is capable of
                     computing a network path or route based on a
                     network graph and applying computational
                     constraints.

   Please note that the terms of CE, PE, and P used throughout this
   document are generic in their definitions.  In particular, whenever
   such acronyms are used, it does not necessarily mean that CE is
   connected to a PE in a VRF environment described in such IETF
   documents as [BGP-MPLSVPN].





Zhang & Vasseur              Informational                      [Page 6]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


3.2.  Objectives and Requirements of Inter-AS Traffic Engineering



   As mentioned in section 1 above, some SPs have requirements for
   achieving the same set of traffic engineering objectives as presented
   in [TE-OVW] across AS boundaries.

   This section examines these requirements in each of the key
   corresponding areas: 1) Inter-AS bandwidth guarantees; 2) Inter-AS
   Resource Optimization and 3) Fast Recovery across ASes, i.e.,
   Recovery of Inter-AS Links/SRLG and ASBR Nodes.

3.2.1.  Inter-AS Bandwidth Guarantees



   The Diffserv IETF working group has defined a set of mechanisms
   described in [DIFF_ARCH], [DIFF_AF], and [DIFF_EF] or [MPLS-Diff].
   These mechanisms can be activated at the edge of or over a Diffserv
   domain to contribute to the enforcement of a QoS policy (or a set of
   QoS policies), which can be expressed in terms of maximum one-way
   transit delay, inter-packet delay variation, loss rate, etc.

   Many SPs have partial or full deployment of Diffserv implementations
   in their networks today, either across the entire network or
   minimally on the edge of the network across CE-PE links.

   In situations where strict QoS bounds are required, admission control
   inside the backbone of a network is in some cases required in
   addition to current Diffserv mechanisms.

   When the propagation delay can be bounded, the performance targets,
   such as maximum one-way transit delay, may be guaranteed by providing
   bandwidth guarantees along the Diffserv-enabled path.

   One typical example of this requirement is to provide bandwidth
   guarantees over an end-to-end path for VoIP traffic classified as EF
   (Expedited Forwarding [DIFF_EF]) class in a Diffserv-enabled network.
   When the EF path is extended across multiple ASes, inter-AS bandwidth
   guarantee is then required.

   Another case for inter-AS bandwidth guarantee is the requirement for
   guaranteeing a certain amount of transit bandwidth across one or
   multiple ASes.

   Several application scenarios are presented to further illustrate
   this requirement in section 4 below.







Zhang & Vasseur              Informational                      [Page 7]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


3.2.2.  Inter-AS Resource Optimization



   In Service Provider (SP) networks, the BGP protocol [BGP] is deployed
   to exchange routing information between ASes.  The inter-AS
   capabilities of BGP may also be employed for traffic engineering
   purposes across the AS boundaries.  Appendix A provides a brief
   description of the current BGP-based inter-AS traffic engineering
   practices.

   SPs have managed to survive with this coarse set of BGP-based traffic
   engineering facilities across inter-AS links in a largely best-effort
   environment.  Certainly, in many cases, ample bandwidth within an
   SP's network and across inter-AS links reduces the need for more
   elaborate inter-AS TE policies.

   However, in the case where a SP network is deployed over multiple
   ASes (for example, as the number of inter-AS links grows), the
   complexity of the inter-AS policies and the difficulty in inter-AS TE
   path optimization increase to a level such that it may soon become
   unmanageable.

   Another example is where inter-AS links are established between
   different SP administrative domains.  Nondeterministic factors such
   as uncoordinated routing and network changes, as well as sub-optimum
   traffic conditions, would potentially lead to a complex set of
   inter-AS traffic engineering policies where current traffic
   engineering mechanisms would probably not scale well.

   In these situations where resource optimization is required and/or
   specific routing requirements arise, the BGP-based inter-AS
   facilities will need to be complemented by a more granular inter-AS
   traffic engineering mechanism.

3.2.3.  Fast Recovery across ASes



   When extending services such as VoIP across ASes, customers often
   require SPs to maintain the same level of performance targets, such
   as packet loss and service availability, as achieved within an AS.
   As a consequence, fast convergence in a stable fashion upon
   link/SRLG/node failures becomes a strong requirement.  This is
   clearly difficult to achieve with current inter-domain techniques,
   especially in cases of link/SRLG failures between ASBRs or ASBR node
   failures.








Zhang & Vasseur              Informational                      [Page 8]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


3.3.  Inter-AS Traffic Engineering Requirements Statement



   Just as in the applicable case of deploying MPLS TE in an SP's
   network, an inter-AS TE method in addition to BGP-based traffic
   engineering capabilities needs to be deployed across inter-AS links
   where resource optimization, bandwidth guarantees and fast recovery
   are required.

   This is especially critical in a Diffserv-enabled, multi-class
   environment described in [PSTE] where statistical performance targets
   must be maintained consistently over the entire path across different
   ASes.

   The approach of extending current intra-AS MPLS TE capabilities
   [TE-RSVP] across inter-AS links for IP/MPLS networks is considered
   here because of already available implementations and operational
   experiences.

   Please note that the inter-AS traffic engineering over an IP-only
   network is for future consideration since there is not sufficient
   interest for similar requirements to those of IP/MPLS networks at
   this time.  More specifically, this document only covers the inter-AS
   TE requirements for packet-based IP/MPLS networks.

4.  Application Scenarios



   The following sections present a few application scenarios over
   IP/MPLS networks where requirements cannot be addressed with the
   current intra-AS MPLS TE mechanism and give rise to considerations
   for inter-AS MPLS traffic engineering requirements.

   Although not explicitly noted in the following discussions, fast
   recovery of traffic path(s) crossing multiple ASes in a stable
   fashion is particularly important in the case of link/SRLG/node
   failures at AS boundaries for all application scenarios presented
   here.

4.1.  Application Scenarios Requiring Inter-AS Bandwidth Guarantees



4.1.1.  Scenario I - Extended or Virtual PoP (VPoP)



   A global service provider (SP1) would like to expand its reach into a
   region where a regional service provider's (SP2) network has already
   established a denser network presence.







Zhang & Vasseur              Informational                      [Page 9]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


   In this scenario, the SP1 may establish interconnections with SP2 in
   one or multiple points in that region.  In their customer-dense
   regions, SP1 may utilize SP2's network as an extended transport by
   co-locating aggregation routers in SP2's PoPs.

   In order to ensure bandwidth capacity provided by SP2 and to achieve
   some degrees of transparency to SP2's network changes in terms of
   capacity and network conditions, one or more inter-AS MPLS TE LSPs
   can be built between SP1's ASBR or PE router inside AS1 and SP1's PE
   routers co-located in SP2's PoPs, as illustrated in the diagram
   below:

    <===========Inter-AS MPLS TE Tunnel===========>
                              -----                -----
                     ________|ASBR |___Inter-AS___|ASBR |________
                    |        | RTR |     Link     | RTR |        |
    ----            -----     -----                -----        -----
   |SP1 |_Inter-AS_| SP2 |                                     | SP1 |
   |VPoP|   Link   |P/PE |                                     |P/PE |
    ----            -----      -----                -----       -----
                     |________|ASBR |___Inter-AS___|ASBR |________|
                              | RTR |     Link     | RTR |
                               -----                -----
    <=================Inter-AS MPLS TE Tunnel======================>
    +-SP1 AS1-+     +---SP2 AS2-----+          +------SP1 AS1------+

   In situations where end-to-end Diffserv paths must be maintained,
   both SPs' networks may need to provision Diffserv PHB at each hop in
   order to support a set of traffic classes with compatible performance
   targets.  The subsequent issues regarding Service Level Agreement
   (SLA) boundaries, reporting and measuring system interoperability and
   support demarcations are beyond the scope of this document and are
   not discussed further.

   If either SP1's or SP2's network is not a Diffserv-aware network, the
   scenario would still apply to provide bandwidth guarantees.

   The SP2, on the other hand, can similarly choose to expand its reach
   beyond its servicing region over SP1's network via inter-AS MPLS TE
   tunnels.

   It is worth mentioning that these remote aggregation routers co-
   located in another SP's network are unlikely to host SP1's IGP and
   BGP routing planes and will more likely maintain their own AS or be
   part of the SP1's AS.  In this case, such TE tunnels may cross
   several ASes, but the Head-end and Tail-end LSRs of TE tunnel may
   have the same AS number, as shown in the diagram above.




Zhang & Vasseur              Informational                     [Page 10]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


4.1.2.  Scenario II - Extended or Virtual Trunk



   Instead of co-locating a PE router in SP2's PoP, SP1 may also choose
   to aggregate customer VPN sites onto a SP2's PE router where inter-AS
   TE tunnels can be built and signaled through SP2's MPLS network
   between the SP2 PoP (to which SP1 and customer CEs are directly
   connected) and SP1's ASBR or PE routers inside SP1's network.  This
   allows SP1's customers connected to SP2 PE router to receive a
   guaranteed bandwidth service up to the TE LSP tail-end router located
   in SP1's network.

   In this scenario, there could be two applicable cases:

   Case 1 - the inter-AS MPLS TE tunnel functions as an extended or
   virtual trunk aggregating SP1's CE's local-loop access circuits on
   SP2's MPLS network over which the bandwidth can be guaranteed to the
   TE LSP tail-end router located in SP1's network, as shown in the
   diagram below:

                        <====Inter-AS MPLS TE Tunnel====>
                                       or
                        < ===Inter-AS MPLS TE Tunnel===============>

    ----               -----     -----                -----     -----
   | CE |_____Local___| SP2 |___|ASBR |___Inter-AS___|ASBR |___|SP1  |
   |    |     Loop    | PE  |   | RTR |     Link     | RTR |   |PE   |
    ----               -----     -----                -----     -----

   +SP1 Customer ASx+ +-----SP2 AS2---+              +-SP1 AS1-------+


   Case 2 - the inter-AS MPLS TE tunnel in this case functions as an
   extended or virtual local access link from SP1's CE on SP2's network
   to the SP1's ASBR or PE:

      <==============Inter-AS MPLS TE Tunnel==============>
                               or
      <==============Inter-AS MPLS TE Tunnel========================>

    ----                -----     -----                -----     -----
   | CE |____Local_____| SP2 |___|ASBR |___Inter-AS___|ASBR |___|SP1  |
   |    |    Loop      | PE  |   | RTR |     Link     | RTR |   |PE   |
    ----                -----     -----                -----     -----

   +SP1 Customer ASx+ +------SP2 AS2---+               +--SP1 AS1-----+






Zhang & Vasseur              Informational                     [Page 11]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


   In Case 2 above, SP2 may elect to establish an aggregating or
   hierarchical intra-AS MPLS TE tunnel between the transiting P or PE
   router and SP2's ASBR router just to reduce the number of tunnel
   states signaled from the SP2 PE to where SP1's CEs are connected.

4.1.3.  Scenario III - End-to-End Inter-AS MPLS TE from CE to CE



   In this scenario as illustrated below, customers require the
   establishment of MPLS TE tunnel from CE1 to CE2 end-to-end across
   several SPs' networks.

    <======================Inter-AS MPLS TE Tunnel==================>

    ---       -----     -----              -----      -----       ---
   |CE1|_____| SP2 |___|ASBR |__Inter-AS__|ASBR |____| SP1 |_____|CE2|
   |   |     | PE  |   | RTR |    Link    | RTR |    | PE  |     |   |
    ---       -----     -----              -----      -----       ---

   +Cust ASx+ +---SP2 AS-----+        +-------SP1 AS-------+ +Cust ASy+

   The diagram below illustrates another example where CE1 and CE2 are
   customers of SP1 with external BGP (eBGP) peering relationships
   established across the CE-PE links.  An inter-AS MPLS TE tunnel may
   then be established from CE1 in ASx to CE2, which may belong to the
   same AS or a different AS than that of CE1 across SP1's network in
   AS2.

    <===============Inter-AS MPLS TE Tunnel=====================>

    ---        -----       ----      ----      -----           ---
   |CE1|______| SP1 |_____|SP1 |____|SP1 |____| SP1 |_________|CE2|
   |   |      | PE1 |     |P1  |    |P2  |    | PE2 |         |   |
    ---        -----       ----      ----      -----           ---

   +-Cust ASx-+ +-------------SP1 AS2----------------+ +-Cust ASy-+

   The above example shows that SP1's network has a single AS.
   Obviously, there may be multiple ASes between CE1 and CE2, as well as
   in the SP1's network.

   In addition, where both CE1 and CE2 reside in the same AS, they will
   likely share the same private AS number.

   However, Scenario III will not scale well if there is a greater
   number of inter-AS TE MPLS tunnels in some degrees of partial mesh or
   full mesh.  Therefore, it is expected that this scenario will have
   few deployments, unless some mechanisms such as hierarchical intra-AS
   TE-LSPs are used to reduce the number of signaling states.



Zhang & Vasseur              Informational                     [Page 12]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


4.2.  Application Scenarios Requiring Inter-AS Resource Optimization



   The scenarios presented in this section mainly deal with inter-AS
   resource optimization.

4.2.1.  Scenario IV - TE across multi-AS within a Single SP
        Administrative Domain



   As mentioned in [TE-APP], SPs have generally admitted that the
   current MPLS TE mechanism provides a great deal of tactical and
   strategic value in areas of traffic path optimization [TE-RSVP] and
   rapid local repair capabilities [TE-FRR] via a set of on-line or
   off-line constraint-based path computation algorithms.

   From a service provider's perspective, another way of stating the
   objectives of traffic engineering is to utilize available capacity in
   the network for delivering customer traffic without violating
   performance targets, and/or to provide better QoS services via an
   improved network utilization, more likely operating below congestion
   thresholds.

   It is worth noting that situations where resource provisioning is not
   an issue (e.g., low density in inter-AS connectivity or ample inter-
   AS capacity), it may not require more scalable and granular TE
   facilities beyond BGP routing policies.  This is because such
   policies can be rather simple and because inter-AS resource
   optimization is not an absolute requirement.

   However many SPs, especially those with networks across multiple
   continents, as well as those with sparsely connected networks, have
   designed their multi-AS routing policies along or within the
   continental or sub-continental boundaries where the number of ASes
   can range from a very few to dozens.  Generally, inter-continent or
   sub-continent capacity is very expensive.  Some Service Providers
   have multiple ASes in the same country and would like to optimize
   resources over their inter-region links.  This would demand a more
   scalable degree of resource optimization, which warrants the
   consideration of extending current intra-AS MPLS TE capabilities
   across inter-AS links.

   In addition, one may only realize higher efficiency in conducting
   traffic optimization and path protection/restoration planning when
   coordinating all network resources as a whole, rather than partially.
   For a network which may consist of many ASes, this could be realized
   via the establishment of inter-AS TE LSPs, as shown in the diagram
   below:





Zhang & Vasseur              Informational                     [Page 13]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


       <===================Inter-AS MPLS Tunnel=============>
     --------                 --------              --------
    |        |_______________|        |____________|        |
    |  SP1   |_______________|  SP1   |____________|  SP1   |
    |  AS1   |_______________|  AS2   |____________|  AS3   |
    |        |               |        |            |        |
     --------                 --------              --------
        ||                                             ||
        ||                   ---------                 ||
        ||___________________|  SP1   |________________||
        |____________________|  AS4   |_________________|
                             |        |
                             ---------

   The motivation for inter-AS MPLS TE is even more prominent in a
   Diffserv-enabled network over which statistical performance targets
   are to be maintained from any point to any point of the network as
   illustrated in the diagram below with an inter-AS DS-TE LSP:

     <===================Inter-AS MPLS DS-TE Tunnel=============>
    ----    -----     -----                -----     -----     ----
   | PE |__| P   |___|ASBR |___Inter-AS___|ASBR |___|P    |___|PE  |
   | RTR|  | RTR |   | RTR |     Link     | RTR |   |RTR  |   |RTR |
    ----    -----     -----                -----     -----     ----
   +------------SP1 AS1---------+        +------------SP1 AS2------+

   For example, the inter-AS MPLS DS-TE LSP shown in the diagram above
   could be used to transport a set of L2 Pseudo Wires or VoIP traffic
   with corresponding bandwidth requirement.

   Furthermore, fast recovery in case of ASBR-ASBR link failure or ASBR
   node failure is a strong requirement for such services.

4.2.2.  Scenario V - Transit ASes as Primary and Redundant Transport



   Scenario V presents another possible deployment case.  SP1 with AS1
   wants to link a regional network to its core backbone by building an
   inter-AS MPLS TE tunnel over one or multiple transit ASes belonging
   to SP2, SP3, etc., as shown in the following diagram:












Zhang & Vasseur              Informational                     [Page 14]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


                <===========Inter-AS MPLS TE Tunnel=======>
   [               ]          [             ]          [              ]
   [  ----    ---- ]          [ ----   ---- ]          [ ----    ---- ]
   [ |P/PE|__|ASBR|]_Inter-AS_[|ASBR|.|ASBR|]_Inter-AS_[|ASBR|  |P/PE|]
   [ |RTR |  |RTR |]   Link   [|RTR | |RTR |]   Link   [|RTR |  |RTR |]
   [  ----    ---- ]          [ ----   ---- ]          [ ----    ---- ]
   [               ]          [             ]          [              ]
       <================Inter-AS MPLS TE Tunnel=====================>
   +SP1 Regional ASx+  +Transit SP2 AS2,etc...SPi ASi+ +------SP1 AS1-+

   This scenario can be viewed as a broader case of Scenario I shown in
   section 4.1.1 where the "VPoP" could be expanded into a regional
   network of SP1.  By the same token, the AS number for SP1's regional
   network ASx may be the same as or different from AS1.

   The inter-AS MPLS TE LSP in this case may also be used to backup an
   internal path, as depicted in the diagram below, although this could
   introduce routing complexities:

                <===========Inter-AS MPLS TE Tunnel=======>
   +----------------------------SP1 AS1-----------------------------+
   [                                                                ]
   [  ----    ----                                     ----    ---- ]
   [ |P/PE|__|ASBR|__________Primary Intera-AS________|P   |  |PE  |]
   [ |RTR |  |RTR |                Link               |RTR |  |RTR |]
   [  ----    ----                                     ----    ---- ]
   [           |                                        |           ]
   [          ----                                     ----         ]
   [         |ASBR|                                   |ASBR|        ]
   [         |RTR |                                   |RTR |        ]
   [          ----                                     ----         ]
               ^ |                                      | ^
               | |                                      | |
               | |            [              ]          | |
               | |            [ ----    ---- ]          | |
               | |__ Inter-AS_[|ASBR|..|ASBR|]_Inter-AS_| |
               |       Link   [|RTR |  |RTR |]   Link     |
               |              [ ----    ---- ]            |
               |              [              ]            |
               |                                          |
               +======Backup Inter-AS MPLS TE Tunnel======+
                 +Transit SP2 AS2,SP3 AS3,etc....SPi ASi+









Zhang & Vasseur              Informational                     [Page 15]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


5.  Detailed Requirements for Inter-AS MPLS Traffic Engineering



   This section discusses detailed requirements for inter-AS MPLS TE in
   two principal areas: 1) requirements for inter-AS MPLS TE in the same
   SP administrative domain and 2) requirements for inter-AS MPLS TE
   across different SP administrative domains.

5.1.  Requirements within One SP Administrative Domain



   This section presents detailed requirements for inter-AS MPLS TE
   within the same SP administrative domain.

5.1.1.  Inter-AS MPLS TE Operations and Interoperability



   The inter-AS MPLS TE solution SHOULD be consistent with requirements
   discussed in [TE-REQ] and the derived solution MUST be such that it
   will interoperate seamlessly with the current intra-AS MPLS TE
   mechanism and inherit its capability sets from [TE-RSVP].

   The proposed solution SHOULD allow the provisioning of a TE LSP at
   the Head/Tail-end with end-to-end Resource Reservation Protocol
   (RSVP) signaling (eventually with loose paths) traversing across the
   interconnected ASBRs, without further provisioning required along the
   transit path.

5.1.2.  Protocol Signaling and Path Computations



   One can conceive that an inter-AS MPLS TE tunnel path signaled across
   inter-AS links consists of a sequence of ASes, ASBRs, and inter-AS
   links.

   The proposed solution SHOULD provide the ability either to select
   explicitly or to auto-discover the following elements when signaling
   the inter-AS TE LSP path:

      - a set of AS numbers as loose hops and/or
      - a set of LSRs including ASBRs

   It should also specify the above elements in the Explicit Route
   Object (ERO) and record them in the Record Route Object (RRO) of the
   Resv message just to keep track of the set of ASes or ASBRs traversed
   by the inter-AS TE LSP.

   In the case of establishing inter-AS TE LSP traversing multiple ASes
   within the same SP networks, the solution SHOULD also allow the
   Head-end LSR to explicitly specify the hops across any one of the
   transiting ASes and the TE tunnel Head-end SHOULD also check the
   explicit segment to make sure that the constraints are met.



Zhang & Vasseur              Informational                     [Page 16]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


   In addition, the proposed solution SHOULD provide the ability to
   specify and signal that certain loose or explicit nodes (e.g., AS
   numbers, etc.) and resources are to be explicitly excluded in the
   inter-AS TE LSP path establishment, such as one defined in
   [EXCLUDE-ROUTE].

5.1.3.  Optimality



   The solution SHOULD allow the set-up of an inter-AS TE LSP that
   complies with a set of TE constraints defined in [TE-REQ]) and
   follows an optimal path.

   An optimal path is defined as a path whose end-to-end cost is
   minimal, based upon either an IGP or a TE metric.  Note that in the
   case of an inter-AS path across several ASes having completely
   different IGP metric policies, the notion of minimal path might
   require IGP metric normalization.

   The solution SHOULD provide mechanism(s) to compute and establish an
   optimal end-to-end path for the inter-AS TE LSP and SHOULD also allow
   for reduced optimality (or sub-optimality) since the path may not
   remain optimal for the lifetime of the LSP.

5.1.4.  Support of Diversely Routed Inter-AS TE LSP



   Setting up multiple inter-AS TE LSPs between a pair of LSRs might be
   desirable when:

     (1) a single TE LSP satisfying the required set of constraints
         cannot be found, in which case it may require load sharing;

     (2) multiple TE paths may be required to limit the impact of a
         network element failure to a portion of the traffic (as an
         example, two VoIP gateways may load balance the traffic among a
         set of inter-AS TE LSPs);

     (3) path protection (e.g., 1:1 or 1:N) as discussed in
         [MPLS-Recov].

   In the examples above, being able to set up diversely routed TE LSPs
   becomes a requirement for inter-AS TE.

   The solution SHOULD be able to set up a set of link/SRLG/Node
   diversely routed inter-AS TE LSPs.







Zhang & Vasseur              Informational                     [Page 17]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


5.1.5.  Re-Optimization



   Once an inter-AS TE LSP has been established, and should there be any
   resource or other changes inside anyone of the ASes, the solution
   MUST be able to re-optimize the LSP accordingly and non-disruptively,
   either upon expiration of a configurable timer or upon being
   triggered by a network event or a manual request at the TE tunnel
   Head-End.

   The solution SHOULD provide an option for the Head-End LSRs to
   control if re-optimizing or not should there exist a more optimal
   path in one of the ASes.

   In the case of an identical set of traversed paths, the solution
   SHOULD provide an option for the Head-End LSRs to control whether
   re-optimization will occur because there could exist a more optimal
   path in one of the transit ASes along the inter-AS TE LSP path.

   Furthermore, the solution MUST provide the ability to reject re-
   optimization at AS boundaries.

5.1.6.  Fast Recovery Support Using MPLS TE Fast Reroute



   There are, in general, two or more inter-AS links between multiple
   pairs of ASBRs for redundancy.  The topological density between ASes
   in a SP network with multi-ASes is generally much higher.  In the
   event of an inter-AS link failure, rapid local protection SHOULD also
   be made available and SHOULD interoperate with the current intra-AS
   MPLS TE fast re-route mechanism from [TE-FRR].

   The traffic routed onto an inter-AS TE tunnel SHOULD also be fast
   protected against any node failure where the node could be internal
   to an AS or at the AS boundary.

5.1.7.  DS-TE Support



   The proposed inter-AS MPLS TE solution SHOULD satisfy core
   requirements documented in [DS-TE].

   It is worth pointing out that the compatibility clause in section 4.1
   of [DS-TE] SHOULD also be faithfully applied to the solution
   development.









Zhang & Vasseur              Informational                     [Page 18]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


5.1.8.  Scalability and Hierarchical LSP Support



   The proposed solution(s) MUST have a minimum impact on network
   scalability from both intra- and inter-AS perspectives.

   This requirement applies to all of the following:

      - IGP (impact in terms of IGP flooding, path computation, etc.)
      - BGP (impact in terms of additional information carried within
        BGP, number of routes, flaps, overload events, etc.)
      - RSVP TE (impact in terms of message rate, number of retained
        states, etc.)

   It is also conceivable that there would potentially be scalability
   issues as the number of required inter-AS MPLS TE tunnels increases.
   In order to reduce the number of tunnel states to be maintained by
   each transiting PoP, the proposed solution SHOULD allow TE LSP
   aggregation such that individual tunnels can be carried onto one or
   more aggregating LSP(s).  One such mechanism, for example, is
   described in [MPLS-LSPHIE].

5.1.9.  Mapping of Traffic onto Inter-AS MPLS TE Tunnels



   There SHOULD be several possibilities to map particular traffic to a
   particular destination onto a specific inter-AS TE LSP.

   For example, static routing could be used if IP destination addresses
   are known.  Another example is to utilize static routing using
   recursive BGP route resolution.

   The proposed solution SHOULD also provide the ability to "announce"
   the inter-AS MPLS TE tunnels as a link into the IGPs (ISIS or OSPF)
   with the link's cost associated with it.  By doing so, PE routers
   that do not participate in the inter-AS TE path computation can take
   into account such links in its IGP-based SPF computation.

5.1.10.  Inter-AS MPLS TE Management



5.1.10.1.  Inter-AS MPLS TE MIB Requirements



   An inter-AS TE Management Information Base (MIB) is required for use
   with network management protocols by SPs to manage and configure
   inter-AS traffic engineering tunnels.  This new MIB SHOULD extend
   (and not reinvent) the existing MIBs to accommodate this new
   functionality.






Zhang & Vasseur              Informational                     [Page 19]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


   An inter-AS TE MIB should have features that include:

      - The setup of inter-AS TE tunnels with associated constraints
        (e.g., resources).
      - The collection of traffic and performance statistics not only at
        the tunnel head-end, but any other points of the TE tunnel.
      - The inclusion of both IPv4/v6 + AS# or AS# subobjects in the ERO
        in the path message, e.g.:

        EXPLICIT_ROUTE class object:
        address1 (loose IPv4 Prefix, /AS1)
        address2 (loose IPv4 Prefix, /AS1)
        AS2      (AS number)
        address3 (loose IPv4 prefix, /AS3)
        address4 (loose IPv4 prefix, /AS3) - destination

        or

        address1 (loose IPv4 Prefix, /AS1)
        address2 (loose IPv4 Prefix, /AS1)
        address3 (loose IPv4 Prefix, /AS2)
        address4 (loose IPv4 Prefix, /AS2)
        address5 (loose IPv4 prefix, /AS3)
        address6 (loose IPv4 prefix, /AS3) - destination

      - Similarly, the inclusion of the RRO object in the Resv message
        recording sub-objects such as interface IPv4/v6 address (if not
        hidden), AS number, a label, a node-id (when required), etc.
      - Inter-AS specific attributes as discussed in section 5 of this
        document including, for example, inter-AS MPLS TE tunnel
        accounting records across each AS segment.

5.1.10.2.  Inter-AS MPLS TE Fault Management Requirements



   In a MPLS network, an SP wants to detect both control plane and data
   plane failures.  But tools for fault detection over LSPs haven't been
   widely developed so far.  SPs today manually troubleshoot such
   failures in a hop-by-hop fashion across the data path.  If they
   detect an error on the data plane, they have to check the control
   plane in order to determine where the faults come from.

   The proposed solution SHOULD be able to interoperate with fault
   detection mechanisms of intra-AS TE and MAY or MAY NOT require the
   inter-AS TE tunnel ending addresses to be known or routable across
   IGP areas (OSPF) or levels (IS-IS) within the transiting ASes with
   working return paths.





Zhang & Vasseur              Informational                     [Page 20]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


   For example, [LSPPING] is being considered as a failure detection
   mechanism over the data plane against the control plane and could be
   used to troubleshoot intra-AS TE LSPs.  Such facilities, if adopted,
   SHOULD then be extended to inter-AS TE paths.

   However, the above example depicts one such mechanism that does
   require a working return path such that diagnostic test packets can
   return via an alternate data plane, such as a global IPv4 path in the
   event that the LSP is broken.

   [MPLS-TTL] presents how TTL may be processed across hierarchical MPLS
   networks, and such a facility as this SHOULD also be extended to
   inter-AS TE links.

5.1.11.  Extensibility



   The solution(s) MUST allow extensions as both inter-AS MPLS TE and
   current intra-AS MPLS TE specifications evolve.

5.1.12.  Complexity and Risks



   The proposed solution(s) SHOULD NOT introduce unnecessary complexity
   to the current operating network to such a degree that it would
   affect the stability and diminish the benefits of deploying such a
   solution over SP networks.

5.1.13.  Backward Compatibility



   The deployment of inter-AS MPLS TE SHOULD NOT impact existing BGP-
   based traffic engineering or MPLS TE mechanisms, but allow for a
   smooth migration or co-existence.

5.1.14.  Performance



   The solution SHOULD be evaluated taking into account various
   performance criteria:

      - Degree of path optimality of the inter-AS TE LSP path
      - TE LSP setup time
      - Failure and restoration time
      - Impact and scalability of the control plane due to added
        overheads, etc.
      - Impact and scalability of the data/forwarding plane due to added
        overheads, etc.







Zhang & Vasseur              Informational                     [Page 21]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


5.2.  Requirements for Inter-AS MPLS TE across Multiple SP
      Administrative Domains



   The requirements for inter-AS MPLS TE across multiple SP admin
   domains SHOULD include all requirements discussed in section 5.1
   above in addition to those that are presented in this section here.

   Please note that the SP with multi-AS networks may choose not to turn
   on the features discussed in the following two sections when building
   TE tunnels across ASes in its own domain.

5.2.1.  Confidentiality



   Since an inter-AS TE LSP may span multiple ASes belonging to
   different SPs, the solution MIGHT allow hiding the set of hops used
   by the TE LSP within an AS, as illustrated in the following example:

   [   ASBR1-----ASBR2   ]
   [       ]     [       ]
   [  A    ]     [   B   ]
   [  AS1  ]     [   AS2 ]
   [  SP1  ]-----[   SP2 ]
   [       ]     [       ]

   Suppose there is an inter-AS TE LSP from A (within AS1 of SP1) to B
   (within AS2 of SP2).  When computing an inter-AS TE LSP path, the set
   of hops within AS2 might be hidden to AS1.  In this case, the
   solution will allow A to learn that the more optimal TE LSP path to B
   (that complies with the set of constraints) traverses ASBR2, without
   a detailed knowledge of the lists of hops used within AS2.

   Optionally, the TE LSP path cost within AS2 could be provided to A
   via, for example, PCC-PCE communication, such that A (PCC) could use
   this information to compute an optimal path, even if the computed
   path is not provided by AS2.  (See [PCE-COM] for PCC-PCE
   communication and [PCE] for a description of the PCE-based path
   computation architecture.)

   In addition, the management requirements discussed in section 5.1.10
   above, when used across different SP admin domains, SHOULD include
   similar confidentiality requirements discussed here in terms of
   "hiding" intermediate hops or interface address and/or labels in the
   transiting or peering SPs.








Zhang & Vasseur              Informational                     [Page 22]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


5.2.2.  Policy Control



   In some cases, policy control might be necessary at the AS
   boundaries, namely ingress policy controls enabling SPs to enforce
   the inter-AS policies per interconnect agreements or to modify some
   requested parameters conveyed by incoming inter-AS MPLS TE signaling
   requests.

   It is worth noting that such a policy control mechanism may also be
   used between ASes within a SP.

   This section discusses only the elements that may be used to form a
   set of ingress control policies, but exactly how SPs establish
   bilateral or multilateral agreements upon which the control policies
   can be built is beyond the scope of this document.

5.2.2.1.  Inter-AS TE Agreement Enforcement Polices



   The following provides a set of TE-LSP parameters in the inter-AS TE
   Requests (RSVP Path Message) that could be enforced at the AS
   boundaries:

      - RSVP-TE session attributes: affinities and preemption priorities
      - Per AS or SP bandwidth admission control to ensure that RSVP-TE
        messages do not request for bandwidth resources over their
        allocation
      - Request origins which can be represented by Head-End tunnel
        ending IP address, originating AS#, neighbor AS#, neighbor ASBR
        interface IP address, etc.
      - DS-TE TE-Class <Class-Type, Preemption>
      - FRR attribute: local protection desired bit, node protection
        desired bit, and bandwidth protection desired bit carried in the
      - SESSION ATTRIBUTE or the FAST-REROUTE objects in the RSVP Path
        message as defined in [TE-FRR]
      - Optimization allowed or not allowed

   In some cases, a TE policy server could also be used for the
   enforcement of inter-AS TE policies.  Implementations SHOULD allow
   the use of a policy enforcement server.  This requirement could allow
   SPs to make the inter-AS TE policies scale better.

   The signaling of a non-policy-compliant request SHOULD trigger the
   generation of a RSVP Path Error message by the policy enforcing node
   towards the Head-end LSR, indicating the cause.  The Head-end LSR
   SHOULD take appropriate actions, such as re-route, upon receipt of
   such a message.





Zhang & Vasseur              Informational                     [Page 23]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


5.2.2.2.  Inter-AS TE Rewrite Policies



   In some situations, SPs may need to rewrite some attributes of the
   incoming inter-AS TE signaling requests due to a lack of resources
   for a particular TE-Class, non-compliant preemption, or mutual
   agreements.  The following provides a non-exhaustive list of the
   parameters that can potentially be rewritten at the AS boundaries:

      - RSVP-TE session attributes: affinities and preemption priorities
      - DS-TE TE-Class <Class-Type, Preemption>
      - ERO expansion requests

   Similarly, the rewriting node SHOULD generate a RSVP Path Error
   Message towards the Head-end LSR indicating the cause in terms of
   types of changes made so as to maintain the end-to-end integrity of
   the inter-AS TE LSP.

5.2.2.3.  Inter-AS Traffic Policing



   The proposed solution SHOULD also provide a set of policing
   mechanisms which could be configured on the inter-AS links to ensure
   that traffic routed through the tunnel does not exceed the bandwidth
   negotiated during LSP signaling.

   For example, an ingress policer could be configured to enforce the
   traffic contract on the mutually agreed resource requirements of the
   established inter-AS TE LSP (i.e., RSVP bandwidth) on the interface
   to which the inter-AS link is connected.

6.  Security Considerations



   The proposed solution(s) MUST address security issues across multiple
   SP administrative domains.  Although inter-AS MPLS TE is not expected
   to add specific security extensions beyond those of current intra-AS
   TE, greater considerations MUST be given in terms of how to establish
   a trusted model across AS boundaries.  SPs SHOULD have a means to
   authenticate (such as using RSVP INTEGRITY Object), to allow, and to
   possibly deny inter-AS signaling requests.  Also, SPs SHOULD be
   protected from DoS attacks.

7.  Acknowledgements



   We would like to thank Yuichi Ikejiri, David Allan, Kurt Erik
   Lindqvist, Dave McDysan, Christian Jacquenet, Kireeti Kompella, Ed
   Kern, Jim Boyle, Thomas Nadeau, Yakov Rekhter, and Bert Wijnen for
   their suggestions and helpful comments during the discussions of this
   document.




Zhang & Vasseur              Informational                     [Page 24]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


8.  Normative References



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

   [TE-RSVP]       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.

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

9.  Informative References



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

   [BGP-MPLSVPN]   Rosen, E. and Y. Rekhter, "BGP/MPLS IP VPNs", Work in
                   Progress, October 2004.

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

   [DIFF_AF]       Heinanen, J., Baker, F., Weiss, W., and J.
                   Wroclawski, "Assured Forwarding PHB Group", RFC 2597,
                   June 1999.

   [DIFF_EF]       Davie, B., Charny, A., Bennet, J.C., Benson, K., Le
                   Boudec, J., Courtney, W., Davari, S., Firoiu, V., and
                   D. Stiliadis, "An Expedited Forwarding PHB (Per-Hop
                   Behavior)", RFC 3246, March 2002.

   [MPLS-Diff]     Le Faucheur, F., Wu, L., Davie, B., Davari, S.,
                   Vaananen, P., Krishnan, R., Cheval, P., and J.
                   Heinanen, "Multi-Protocol Label Switching (MPLS)
                   Support of Differentiated Services", RFC 3270, May
                   2002.

   [TE-OVW]        Awduche, D., Chiu, A., Elwalid, A., Widjaja, I., and
                   X. Xiao, "Overview and Principles of Internet Traffic
                   Engineering", RFC 3272, May 2002.

   [PSTE]          Li, T. and Y. Rekhter, "A Provider Architecture for
                   Differentiated Services and Traffic Engineering
                   (PASTE)", RFC 2430, October 1998.



Zhang & Vasseur              Informational                     [Page 25]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


   [TE-APP]        Boyle, J., Gill, V., Hannan, A., Cooper, D., Awduche,
                   D., Christian, B., and W. Lai, "Applicability
                   Statement for Traffic Engineering with MPLS", RFC
                   3346, August 2002.

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

   [BGP]           Rekhter, Y. and T. Li, "A Border Gateway Protocol 4
                   (BGP-4)", RFC 1771, March 1995.

   [LSPPING]       Kompella, K. and G. Swallow, "Detecting MPLS Data
                   Plane Failures", Work in Progress, May 2005.

   [MPLS-TTL]      Agarwal, P. and B. Akyol, "Time To Live (TTL)
                   Processing in Multi-Protocol Label Switching (MPLS)
                   Networks", RFC 3443, January 2003.

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

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

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

   [MPLS-Recov]    Sharma, V. and F. Hellstrand, "Framework for Multi-
                   Protocol Label Switching (MPLS)-based Recovery", RFC
                   3469, February 2003.

   [EXCLUDE-ROUTE] Lee, CY., Farrel, A., and S. De Cnodder, "Exclude
                   Routes - Extension to RSVP-TE", Work in Progress,
                   August 2005.

   [PCE]           Farrel, A., Vasseur, J.-P., and J. Ash, "Path
                   Computation Element (PCE) Architecture", Work in
                   Progress, September 2005.

   [PCE-COM]       Vasseur, J.-P., et al., "Path Computation Element
                   (PCE) communication Protocol (PCEP) - Version 1",
                   Work in Progress, September 2005.



Zhang & Vasseur              Informational                     [Page 26]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


Appendix A.  Brief Description of BGP-based Inter-AS Traffic
             Engineering



   In today's Service Provider (SP) network, BGP is deployed to meet two
   different sets of requirements:

      - Establishing a scalable exterior routing plane separate from the
        data forwarding plane within SP's administrative domain
      - Exchanging network reachability information with different BGP
        autonomous systems (ASes) that could belong to a different SP or
        simply, a different AS within a SP network

   Over connections across the AS boundaries, traffic engineering may
   also be accomplished via a set of BGP capabilities by appropriately
   enforcing BGP-based inter-AS routing policies.  The current BGP-based
   inter-AS traffic engineering practices may be summarized as follows:

      - "Closest exit" routing where egress traffic from one SP to
        another follows the path defined by the lowest IGP or intra-AS
        MPLS TE tunnel metrics of the BGP next-HOP of exterior routes
        learned from other ASes over the inter-AS links
      - "BGP path attribute"-based routing selection mechanism where the
        egress traffic path is determined by interconnect (peering or
        transit) policies based upon one or a combination of BGP path
        attributes, like AS_PATH, MULTI_EXIT_DISC (MED), and Local_Pref.

   SPs have often faced a number of nondeterministic factors in the
   practices of inter-AS traffic engineering employing the methods
   mentioned above:

      - Sub-optimum traffic distribution across inter-AS links
      - Nondeterministic traffic condition changes due to uncoordinated
        IGP routing policies or topology changes within other AS and
        uncoordinated BGP routing policy changes (MED or as-prepend,
        etc.)

   In addition, to achieve some degrees of granularity, SPs may choose
   to enforce BGP inter-AS policies.  These policies are specific to one
   inter-AS link or to a set of inter-AS links for ingress traffic.  By
   tagging certain sets of routes with a specific attribute when
   announcing to another AS, the ingress traffic is destined to certain
   PoPs or to regions within SP's network from another AS.  Of course,
   this operates on the assumption that the other AS permits automated
   egress policy by matching the predefined attribute from incoming
   routes.






Zhang & Vasseur              Informational                     [Page 27]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


Editors' Addresses



   Raymond Zhang
   Infonet Services Corporation
   2160 E. Grand Ave.
   El Segundo, CA 90025
   USA

   EMail: raymond_zhang@infonet.com


   J.-P. Vasseur
   Cisco Systems, Inc.
   300 Beaver Brook Road
   Boxborough, MA 01719
   USA

   EMail: jpv@cisco.com

































Zhang & Vasseur              Informational                     [Page 28]

RFC 4216             MPLS Inter-AS TE Requirements         November 2005


Full Copyright Statement



   Copyright (C) The Internet Society (2005).

   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
   "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS
   OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE INTERNET
   ENGINEERING TASK FORCE DISCLAIM 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.

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
   http://www.ietf.org/ipr.

   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 ietf-
   ipr@ietf.org.

Acknowledgement



   Funding for the RFC Editor function is currently provided by the
   Internet Society.







Zhang & Vasseur              Informational                     [Page 29]