idnits 2.17.1 draft-ietf-pce-inter-layer-frwk-02.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** It looks like you're using RFC 3978 boilerplate. You should update this to the boilerplate described in the IETF Trust License Policy document (see https://trustee.ietf.org/license-info), which is required now. -- Found old boilerplate from RFC 3978, Section 5.1 on line 20. -- Found old boilerplate from RFC 3978, Section 5.5 on line 872. -- Found old boilerplate from RFC 3979, Section 5, paragraph 1 on line 846. -- Found old boilerplate from RFC 3979, Section 5, paragraph 3 on line 861. ** This document has an original RFC 3978 Section 5.4 Copyright Line, instead of the newer IETF Trust Copyright according to RFC 4748. ** This document has an original RFC 3978 Section 5.5 Disclaimer, instead of the newer disclaimer which includes the IETF Trust according to RFC 4748. ** The document seems to lack an RFC 3979 Section 5, para. 2 IPR Disclosure Acknowledgement -- however, there's a paragraph with a matching beginning. Boilerplate error? Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- ** The document seems to lack a 1id_guidelines paragraph about 6 months document validity -- however, there's a paragraph with a matching beginning. Boilerplate error? == The page length should not exceed 58 lines per page, but there was 15 longer pages, the longest (page 1) being 60 lines Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- ** The document seems to lack an IANA Considerations section. (See Section 2.2 of https://www.ietf.org/id-info/checklist for how to handle the case when there are no actions for IANA.) ** There are 15 instances of too long lines in the document, the longest one being 1 character in excess of 72. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the RFC 3978 Section 5.4 Copyright Line does not match the current year == The document seems to lack the recommended RFC 2119 boilerplate, even if it appears to use RFC 2119 keywords. (The document does seem to have the reference to RFC 2119 which the ID-Checklist requires). -- The document seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (October 2006) is 6374 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- == Missing Reference: 'RFC3031' is mentioned on line 89, but not defined == Unused Reference: 'RFC2119' is defined on line 770, but no explicit reference was found in the text == Unused Reference: 'RFC4208' is defined on line 780, but no explicit reference was found in the text == Unused Reference: 'RFC4657' is defined on line 790, but no explicit reference was found in the text == Unused Reference: 'RFC4674' is defined on line 794, but no explicit reference was found in the text == Unused Reference: 'PCEP' is defined on line 807, but no explicit reference was found in the text Summary: 7 errors (**), 0 flaws (~~), 9 warnings (==), 6 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group Eiji Oki 3 Internet Draft NTT 4 Category: Informational Jean-Louis Le Roux 5 Expires: April 2007 France Telecom 6 Adrian Farrel 7 Old Dog Consulting 8 October 2006 10 Framework for PCE-Based Inter-Layer MPLS and GMPLS Traffic 11 Engineering 13 draft-ietf-pce-inter-layer-frwk-02.txt 15 Status of this Memo 17 By submitting this Internet-Draft, each author represents that any 18 applicable patent or other IPR claims of which he or she is aware 19 have been or will be disclosed, and any of which he or she becomes 20 aware will be disclosed, in accordance with Section 6 of BCP 79. 22 Internet-Drafts are working documents of the Internet Engineering 23 Task Force (IETF), its areas, and its working groups. Note that 24 other groups may also distribute working documents as Internet- 25 Drafts. 27 Internet-Drafts are draft documents valid for a maximum of six 28 months and may be updated, replaced, or obsoleted by other 29 documents at any time. It is inappropriate to use Internet- Drafts 30 as reference material or to cite them other than as "work in 31 progress." 33 The list of current Internet-Drafts can be accessed at 34 http://www.ietf.org/ietf/1id-abstracts.txt. 36 The list of Internet-Draft Shadow Directories can be accessed at 37 http://www.ietf.org/shadow.html. 39 Abstract 41 A network may comprise of multiple layers. It is important to 42 globally optimize network resources utilization, taking into 43 account all layers, rather than optimizing resource utilization at 44 each layer independently. This allows better network efficiency to 45 be achieved through a process that we call inter-layer traffic 46 engineering. The Path Computation Element (PCE) can be a powerful 47 tool to achieve inter-layer traffic engineering. 49 This document describes a framework for applying the PCE-based path 50 computation architecture to inter-layer MPLS and GMPLS traffic 51 engineering. It provides suggestions for the deployment of PCE in 52 support of multi-layer networks. This document also describes 53 network models where PCE performs inter-layer traffic engineering, 54 and the relationship between PCE and a functional component called 55 the Virtual Network Topology Manager (VNTM). 57 Table of Contents 59 1. Terminology.....................................................2 60 2. Introduction....................................................2 61 3. Inter-Layer Path Computation....................................3 62 4. Inter-layer Path Computation Models.............................5 63 4.1. Single PCE Inter-Layer Path Computation......................5 64 4.2. Multiple PCE Inter-Layer Path Computation....................6 65 4.3. General observations.........................................7 66 5. Inter-Layer Path Control........................................7 67 5.1. VNT Management...............................................8 68 5.2. Inter-Layer Path Control Models..............................8 69 5.2.1. Cooperation model between PCE and VNTM.....................8 70 5.2.2. Higher-Layer Signaling Trigger Model......................10 71 5.2.3. Examples of multi-layer ERO...............................12 72 6. Choosing between inter-layer path control models...............12 73 6.1. VNTM functions:.............................................13 74 6.2. Border LSR functions:.......................................13 75 6.3. Complete inter-layer LSP setup time:........................13 76 6.4. Network complexity..........................................14 77 7. Security Considerations........................................14 78 8. Acknowledgment.................................................15 79 9. References.....................................................15 80 9.1. Normative Reference.........................................15 81 9.2. Informative Reference.......................................15 82 10. Authors' Addresses...........................................16 83 11. Intellectual Property Statement..............................16 85 1. Terminology 87 This document uses terminology from the PCE-based path computation 88 architecture [RFC4655] and also common terminology from Multi 89 Protocol Label Switching (MPLS) [RFC3031], Generalized MPLS (GMPLS) 90 [RFC3945] and Multi-Layer Networks [MLN-REQ]. 92 2. Introduction 94 A network may comprise of multiple layers. These layers may 95 represent separations of technologies (e.g., packet switch capable 96 (PSC), time division multiplex (TDM) lambda switch capable (LSC)) 97 [RFC3945], separation of data plane switching granularity levels 98 (e.g. PSC-1, PSC-2, VC4, VC12) [MLN-REQ], or a distinction between 99 client and server networking roles. In this multi-layer network, 100 Label Switched Paths (LSPs) in a lower layer are used to carry 101 higher-layer LSPs across the lower-layer network. The network 103 Oki et al Expires April 2007 2 104 topology formed by lower-layer LSPs and advertised to the higher 105 layer is called a Virtual Network Topology (VNT) [MLN-REQ]. 107 It may be effective to optimize network resource utilization 108 globally, i.e. taking into account all layers, rather than 109 optimizing resource utilization at each layer independently. This 110 allows better network efficiency to be achieved and is what we call 111 inter-layer traffic engineering. This includes mechanisms allowing 112 the computation of end-to-end paths across layers (known as inter- 113 layer path computation), and mechanisms for control and management 114 of the VNT by setting up and releasing LSPs in the lower layers 115 [MLN-REQ]. 117 Inter-layer traffic engineering is included in the scope of the 118 PCE-based path computation architecture [RFC4655], and PCE can 119 provide a suitable mechanism for resolving inter-layer path 120 computation issues. 122 PCE Communication Protocol requirements for inter-layer traffic 123 engineering are set forth in [PCE-INTER-LAYER-REQ]. 125 This document describes a framework for applying the PCE-based path 126 computation architecture to inter-layer traffic engineering. It 127 provides suggestions for the deployment of PCE in support of multi- 128 layer networks. This document also describes network models where 129 PCE performs inter-layer traffic engineering, and the relationship 130 between PCE and a functional component in charge of the control and 131 management of the VNT, and called the Virtual Network Topology 132 Manager (VNTM). 134 3. Inter-Layer Path Computation 136 This section describes key topics of inter-layer path computation 137 in MPLS and GMPLS networks. 139 [RFC4206] defines a way to signal a higher-layer LSP, whose 140 explicit route includes hops traversed by LSPs in lower layers. The 141 computation of end-to-end paths across layers is called Inter-Layer 142 Path Computation. 144 A Label Switching Router (LSR) in the higher-layer might not have 145 information on the lower-layer topology, particularly in an overlay 146 or augmented model, and hence may not be able to compute an end-to- 147 end path across layers. 149 PCE-based inter-layer path computation, consists of relying on one 150 or more PCEs to compute an end-to-end path across layers. This 151 could be achieved by rely on a single PCE path computation where 152 the PCE has topology information about multiple layers and can 153 directly compute an end-to-end path across layers considering the 154 topology of all of the layers. Alternatively, the inter-layer path 155 computation could be performed as a multiple PCE computation where 156 each member of a set of PCEs have information about the topology of 158 Oki et al Expires April 2007 3 159 one or more layers, but not all layers, and collaborate to compute 160 an end-to-end path. 162 Consider, for instance, a two-layer network where the higher-layer 163 network is a packet-based IP/MPLS network or GMPLS network and the 164 lower-layer network is a GMPLS optical network. An ingress LSR in 165 the higher-layer network tries to set up an LSP to an egress LSR 166 also in the higher-layer network across the lower-layer network, 167 and needs a path in the higher-layer network. However, suppose that 168 there is no Traffic Engineering (TE) link between border LSRs, 169 which are located on the boundary between the higher-layer and 170 lower-layer networks, and that the ingress LSR does not have 171 topology visibility in the lower layer. If a single-layer path 172 computation is applied for the higher-layer, the path computation 173 fails. On the other hand, inter-layer path computation is able to 174 provide a route in the higher-layer and a suggestion that a lower- 175 layer LSP be setup between border LSRs, considering both layers' TE 176 topologies. 178 Lower-layer LSPs form a Virtual Network Topology (VNT), which can 179 be used for routing higher-layer LSPs or to carry IP traffic. 180 Inter-layer path computation for end-to-end LSPs in the higher- 181 layer network that span the lower-layer network may utilize the VNT, 182 and PCE is a candidate for computing the paths of such higher-layer 183 LSPs within the higher-layer network. The PCE-based path 184 computation model can: 186 - Perform a single computation on behalf of the ingress LSR using 187 information gathered from more than one layer. This mode is 188 referred to as Single PCE Computation in [RFC4655]. 190 - Compute a path on behalf of the ingress LSR through cooperation 191 between PCEs responsible for each layer. This mode is referred to 192 as Multiple PCE Computation with inter-PCE communication in 193 [RFC4655]. 195 - Perform separate path computations on behalf of the TE-LSP head- 196 end and each transit LSR that is the entry point to a new layer. 197 This mode is referred to as Multiple PCE Computation (without 198 inter-PCE communication) in [RFC4655]. This option utilizes per- 199 layer path computation performed independently by successive PCEs. 201 The PCE computes and returns a path to the PCC that the PCC can use 202 to build an MPLS or GMPLS LSP once converted to an Explicit Route 203 Object (ERO) for use in RSVP-TE signaling. There are two options. 205 - Option 1: Mono-layer path. 206 The PCE computes a "mono-layer" path, i.e., a path that includes 207 only TE links from the same layer. There are two cases for this 208 option. In the first case the PCE computes a path that includes 209 already established lower-layer LSPs or expected lower-layer LSPs 210 to be established: that is the resulting ERO includes sub-object(s) 211 corresponding to lower-layer hierarchical LSPs expressed as the TE 212 link identifiers, which can be numbered or unnumbered ones, of the 214 Oki et al Expires April 2007 4 215 hierarchical LSPs when advertised as TE links in the higher-layer 216 network. The TE link may be a regular TE link that is actually 217 established, or a virtual TE link that is not established yet (see 218 [MLN-REQ]). If it is a regular TE link, this does not trigger new 219 lower-layer LSP setup, but the utilization of existing lower-layer 220 LSPs. If it is a virtual TE link, this triggers a new lower-layer 221 LSP setup (provided that there are available resources in the lower 222 layer). A transit LSR corresponding to the entry point of the 223 virtual TE link is expected to trigger the new lower-layer LSP 224 setup. Note that the path of a virtual TE link is not necessarily 225 known in advance, and this may require path computation either on 226 the entry point or on a PCE. The second case is that the PCE 227 computes a path that includes loose hop(s). The higher layer path 228 would select which lower layer paths to use and would select the 229 entry and exit points from those layers, but would not select the 230 path across the layers. A transit LSR corresponding to the entry 231 point is expected to expand the loose hop (either itself or relying 232 on the services of a PCE). The path expansion process on the border 233 LSR may result either in the selection of an existing lower-layer 234 LSP, or in the computation and setup of a new lower-layer LSP. 236 - Option 2: Multi-layer path. The PCE computes a "multi-layer" path, 237 i.e. a path that includes TE links from distinct layers [RFC4206]. 238 Such a path can include the complete path of one or more lower- 239 layer LSPs that already exist or are not yet established. In the 240 latter case, the signaling of the higher-layer LSP will trigger the 241 establishment of the lower-layer LSPs. 243 4. Inter-layer Path Computation Models 245 As stated in Section 3, two PCE modes defined in the PCE 246 architecture can be used to perform inter-layer path computation. 247 They are discussed below. 249 4.1. Single PCE Inter-Layer Path Computation 251 In this model Inter-layer path computation is performed by a single 252 PCE that has topology visibility in all layers. Such a PCE is 253 called a multi-layer PCE. 255 In Figure 1, the network is comprised of two layers. LSR H1, H2, H3 256 and H4 belong to the higher layer, and LSRs L1 and L2 belong to the 257 lower layer. The PCE is a multi-layer PCE that has visibility into 258 both layers. It can perform end-to-end path computation across 259 layers (single PCE path computation). For instance, it can compute 260 an optimal path H2-L1-L2-H3-H4, for a higher layer LSP from H1 to 261 H4. This path includes the path of a lower layer LSP from H2 to H3, 262 already established or not. 264 Oki et al Expires April 2007 5 265 ----- 266 | PCE | 267 ----- 268 ----- ----- ----- ----- 269 | LSR |--| LSR |................| LSR |--| LSR | 270 | H1 | | H2 | | H3 | | H4 | 271 ----- -----\ /----- ----- 272 \----- -----/ 273 | LSR |--| LSR | 274 | L1 | | L2 | 275 ----- ----- 277 Figure 1 : Multi-Layer PCE - A single PCE with multi-layer 278 visibility 280 4.2. Multiple PCE Inter-Layer Path Computation 282 In this model there is at least one PCE per layer, and each PCE has 283 topology visibility restricted to its own layer. Some providers may 284 want to keep the layer boundaries due to other factors such as 285 organizational and/or service management issues. The choice for 286 multiple PCE computation instead of single PCE computation may also 287 be driven by scalability considerations, as in this mode a PCE only 288 needs to maintain topology information of one layer (TED size 289 reduction). 291 These PCEs are called mono-layer PCEs. Mono-layer PCEs collaborate 292 to compute an end-to-end optimal path across layers. 294 In Figure 2, there is one PCE in each layer. The PCEs from each 295 layer collaborate to compute an end-to-end path across layers. PCE 296 Hi is responsible for computations in the higher layer and may 297 "consult" with PCE Lo to compute paths across the lower layer. PCE 298 Lo is responsible for path computation in the lower layer. A simple 299 example of cooperation between the PCEs could be: PCE Hi requests a 300 path H2-H3 from PCE Lo. Of course more complex cooperation may be 301 required if an end-to-end optimal path is desired. 303 Oki et al Expires April 2007 6 304 ----- 305 | PCE | 306 | Hi | 307 --+-- 308 | 309 ----- ----- | ----- ----- 310 | LSR |--| LSR |............|...........| LSR |--| LSR | 311 | H1 | | H2 | | | H3 | | H4 | 312 ----- -----\ --+-- /----- ----- 313 \ | PCE | / 314 \ | Lo | / 315 \ ----- / 316 \ / 317 \----- -----/ 318 | LSR |--| LSR | 319 | L1 | | L2 | 320 ----- ----- 322 Figure 2 : Cooperating Mono-Layer PCEs - Multiple PCEs with single- 323 layer visibility 325 4.3. General observations 327 - Depending on implementation details, inter-layer path computation 328 time in the Single PCE inter-layer path computation model may be 329 less than that of the Multiple PCE model with cooperating mono- 330 layer PCEs, because there is no requirement to exchange messages 331 between cooperating PCEs. 333 - When TE topology for all layered networks is visible within one 334 routing domain, the single PCE inter-layer path computation model 335 may be adopted because a PCE is able to collect all layers' TE 336 topologies by participating in only one routing domain. 338 - As the single PCE inter-layer path computation model uses more TE 339 topology information than is used by PCEs in the Multiple PCE path 340 computation model, it requires more computation power and memory. 342 When there are multiple candidate layer border nodes (we may say 343 that the higher layer is multi-homed), optimal path computation 344 requires that all the possible paths transiting different layer 345 border nodes or links be examined. This is relatively simple in the 346 single PCE inter-layer path computation model because the PCE has 347 full visibility - the computation is similar to the computation 348 within a single domain of a single layer. In the multiple PCE 349 inter-layer path computation model, backward recursive techniques 350 described in [BRPC] could be used, by considering layers as 351 separate domains. 353 5. Inter-Layer Path Control 355 Oki et al Expires April 2007 7 357 5.1. VNT Management 359 As a result of inter-layer path computation, a PCE may determine 360 that there is insufficient bandwidth available in the higher-layer 361 network to support this or future higher-layer LSPs. The problem 362 might be resolved if new LSPs are provisioned across the lower- 363 layer network. Further, the modification, re-organization and new 364 provisioning of lower-layer LSPs may enable better utilization of 365 lower-layer network resources given the demands of the higher-layer 366 network. In other words, the VNT needs to be controlled or managed 367 in cooperation with inter-layer path computation. 369 A VNT Manager (VNTM) is defined as a network element that manages 370 and controls the VNT. PCE and "VNT Management" are distinct 371 functions that may or may not be co-located. To describe each 372 function clearly, VNTM is considered as a functional element in 373 this draft. 375 5.2. Inter-Layer Path Control Models 377 5.2.1. Cooperation model between PCE and VNTM 379 ----- ------ 380 | PCE |--->| VNTM | 381 ----- ------ 382 ^ : 383 : : 384 : : 385 v V 386 ----- ----- ----- ----- 387 | LSR |----| LSR |................| LSR |----| LSR | 388 | H1 | | H2 | | H3 | | H4 | 389 ----- -----\ /----- ----- 390 \----- -----/ 391 | LSR |--| LSR | 392 | L1 | | L2 | 393 ----- ----- 395 Figure 3: Cooperation model between PCE and VNTM 397 A multi-layer network consists of higher-layer and lower-layer 398 networks. LSRs H1, H2, H3, and H4 belong to the higher-layer 399 network, LSRs H2, L1, L2, and H3 belong to the lower-layer network, 400 as shown in Figure 3. Consider that H1 requests PCE to compute an 401 inter-layer path between H1 and H4. There is no TE link in the 402 higher-layer between H2 and H3 before the path computation request. 404 The roles of PCE and VNTM are as follows. PCE performs inter-layer 405 path computation and is unable to supply a path because there is no 406 TE link between H2 and H3. The computation fails, but PCE suggests 407 to VNTM that a lower-layer LSP (H2-H3) should be established to 408 support future LSP requests. VNTM uses local policy and possibly 409 management/configuration input to determine how to process the 410 suggestion from PCE, and may request an ingress LSR (e.g. H2) to 412 Oki et al Expires April 2007 8 413 establish a lower-layer LSP. VNTM or the ingress LSR (H2) may use a 414 PCE with visibility into the lower layer to compute the path of 415 this new LSP. 417 If the PCE cannot compute a path for the higher-layer LSP without 418 the establishment of a further lower-layer LSP, the PCE may notify 419 VNTM and wait for the lower-layer LSP to be set up and advertised 420 as a TE link. It can then compute the complete end-to-end path for 421 the higher-layer LSP and return the result to the PCC. In this case, 422 the PCC may be kept waiting some time, and it is important that the 423 PCC understands this. It is also important that the PCE and VNTM 424 have an agreement that the lower-layer LSP will be set up in a 425 timely manner, the PCE operates a timeout, or the PCE will be 426 notified by VNTM that no new LSP will become available. An example 427 of such a cooperative procedure between PCE and VNTM is as follows. 429 Step 1: H1 (PCC) requests PCE to compute a path between H1 and H4. 430 In the request, it indicates that inter-layer path computation is 431 allowed. 433 Step 2: As a result of the inter-layer path computation, PCE judges 434 that a new lower-layer LSP needs to be established. 436 Step 3: PCE suggests to VNTM that a new lower-layer LSP should be 437 established if necessary and if acceptable within VNTM's policy 438 constraints. The inter-layer path route computed by PCE may include 439 one or more virtual TE links. If PCE knows the inclusion of the 440 virtual TE link(s) in the inter-layer route, PCE may suggest VNTM 441 that the corresponding new lower-layer LSP(s) should be established. 442 Otherwise, new lower-layer LSP(s) may be setup according to the 443 higher-layer signaling trigger model. 445 In the above description, it is assumed that a higher layer LSP is 446 supported by a single lower layer LSP. However, in case of VCAT, 447 several lower layer LSPs may be used to transport a single higher 448 layer LSP. 450 Step 4: VNTM requests an ingress LSR (e.g. H2) to establish a 451 lower-layer LSP. The request message may include a pre-computed 452 lower-layer LSP route obtained from the PCE responsible for the 453 lower-layer network. 455 Step 5: The ingress LSR starts signaling to establish a lower-layer 456 LSP. 458 Step 6: If the lower-layer LSP setup is completed, the ingress LSR 459 notifies VNTM that the LSP is complete and supplies the tunnel 460 information. 462 Step 7: VNTM replies to PCE to inform it that the lower-layer LSP 463 is now established, and includes the lower-layer tunnel information. 464 Alternatively, PCE may get to know about the existence of the 465 lower-layer LSP when a new TE link in the higher-layer 467 Oki et al Expires April 2007 9 468 corresponding to the lower-layer LSP is advertised to PCE through 469 the IGP. 471 Step 8: PCE replies to H1 (PCC) with a computed higher-layer LSP 472 route. The computed path is categorized as a mono-layer path that 473 includes the already-established lower layer-LSP. The higher-layer 474 route is specified as H2-H3-H4, where all hops are strict. 476 Step 9: H1 initiates signaling with the computed path H2-H3-H4 to 477 establish the higher-layer LSP. 479 5.2.2. Higher-Layer Signaling Trigger Model 481 ----- 482 | PCE | 483 ----- 484 ^ 485 : 486 : 487 v 488 ----- ----- ----- ----- 489 | LSR |----| LSR |................| LSR |--| LSR | 490 | H1 | | H2 | | H3 | | H4 | 491 ----- -----\ /----- ----- 492 \----- -----/ 493 | LSR |--| LSR | 494 | L1 | | L2 | 495 ----- ----- 497 Figure 4: Higher-layer signaling trigger model 499 Figure 4 shows the higher-layer signaling trigger model. As in the 500 case described in section 5.2.1, consider that H1 requests PCE to 501 compute an inter-layer path between H1 and H4. There is no TE link 502 in the higher-layer between H2 and H3 before the path computation 503 request. 505 If PCE judges that a lower-layer LSP needs to be established based 506 on the inter-layer path computation result, a lower-layer LSP is 507 established during the higher-layer signaling procedure. After PCE 508 completes inter-layer path computation, PCE sends a reply message 509 including explicit route to the ingress LSR (PCC). There are two 510 ways to express the higher-layer LSP route, which are a multi-layer 511 path and a mono-layer path that includes loose hop(s). 513 In the higher-layer signaling trigger model with a multi-layer path, 514 a high-layer LSP route includes a route for a lower-layer LSP that 515 is not yet established. An LSR that is located at the boundary 516 between the higher-layer and lower-layer networks, called a border 517 LSR, receives a higher-layer signaling message and then may start 518 to setup the lower-layer LSP. Note that it depends on a policy at 519 the border LSR whether the higher-layer signaling triggers a lower- 520 layer LSP setup. An example procedure of the signaling trigger 521 model with a multi-layer path is as follows. 523 Oki et al Expires April 2007 10 524 Step 1: H1 (PCC) requests PCE to compute a path between H1 and H4. 525 The request indicates that inter-layer path computation is allowed. 527 Step 2: As a result of the inter-layer path computation, PCE judges 528 that a new lower-layer LSP needs to be established. 530 Step 3: PCE replies to H1 (PCC) with a computed multi-layer route 531 including higher-layer and lower-layer LSP routes. The route may be 532 specified as H2-L1-L2-H3-H4, where all hops are strict. 534 Step 4: H1 initiates higher-layer signaling using the computed 535 explicit router of H2-L1-L2-H3-H4. 537 Step 5: The border LSR (H2) that receives the higher-layer 538 signaling message starts lower-layer signaling to establish a 539 lower-layer LSP along the specified lower-layer route of L1-L2-H3. 540 That is, the border LSR recognizes the hops within the explicit 541 route that apply to the lower-layer network, verifies with local 542 policy that a new LSP is acceptable, and establishes the required 543 lower-layer LSP. Note that it is possible that a suitable lower- 544 layer LSP has been established (or become available) between the 545 time that the computation was performed and the moment when the 546 higher-layer signaling message reached the border LSR. In this case, 547 the border LSR may select such a lower-layer LSP without the need 548 to signal a new LSP provided that the lower-layer LSP satisfies the 549 explicit route in the higher-layer signaling request. 551 Step 6: After the lower-layer LSP is established, the higher-layer 552 signaling continues along the specified higher-layer route of H2- 553 H3-H4. 555 On the other hand, in the signaling trigger model with a mono-layer 556 path, a higher-layer LSP route includes a loose or strict hop to 557 traverse the lower-layer network between the two border LSRs. In 558 the strict hop case, a virtual TE link may be advertised, but a 559 lower-layer LSP is not setup. A border LSR that receives a higher- 560 layer signaling message needs to determine a path for a new lower- 561 layer LSP. It applies local policy to verify that a new LSP is 562 acceptable and then either consults a PCE with responsibility for 563 the lower-layer network or computes the path by itself, and 564 initiates signaling to establish a lower-layer LSP. Again, it is 565 possible that a suitable lower-layer LSP has been established (or 566 become available) between the time that the higher-layer 567 computation was performed and the moment when the higher-layer 568 signaling message reached the border LSR. In this case, the border 569 LSR may select such a lower-layer LSP without the need to signal a 570 new LSP provided that the lower-layer LSP satisfies the explicit 571 route in the higher-layer signaling request. Since the higher-layer 572 signaling request used a loose hop without specifying any specifics 573 of the path within the lower-layer network, the border LSR has 574 greater freedom to choose a lower-layer LSP than in the previous 575 example. 577 Oki et al Expires April 2007 11 578 The difference between procedures of the signaling trigger model 579 with a multi-layer path and a mono-layer path is Step 5. Step 5 of 580 the signaling trigger model with a mono layer path is as follows: 582 Step 5' The border LSR (H2) that receives the higher-layer 583 signaling message applies local policy to verify that a new LSP is 584 acceptable and then initiates establishment of a lower-layer LSP. 585 It either consults a PCE with responsibility for the lower-layer 586 network or computes the route by itself to expand the loose hop 587 route in the higher-layer path. 589 5.2.3. Examples of multi-layer ERO 591 PCE 592 ^ 593 : 594 : 595 V 596 H1--H2 H3--H4 597 \ / 598 L1==L2==L3--L4--L5 599 | 600 | 601 L6--L7 602 \ 603 H5--H6 605 Figure 5 Example of multi-layer network 607 This section describes how lower-layer LSP setup is performed in 608 the higher-layer signaling trigger model using an ERO that can 609 include subobjects in both the higher and lower layers. It gives 610 rise to several options for the ERO when it reaches the last LSR in 611 the higher layer network (H2). 612 1. The next subobject is a loose hop to H3 (mono layer ERO). 613 2. The next subobject is a strict hop to L1 followed by a loose hop 614 to H3. 615 3. The next subobjects are a series of hops (strict or loose) in 616 the lower-layer network followed by H3. For example, {L1(strict), 617 L3(loose), L5(loose), H3(strict)} 619 In the first, the lower layer can utilize any LSP tunnel that will 620 deliver the end-to-end LSP to H3. In the third case, the lower 621 layer must select an LSP tunnel that traverses L3 and L5. However, 622 this does not mean that the lower layer can or should use an LSP 623 from L1 to L3 and another from L3 to L5. 625 6. Choosing between inter-layer path control models 627 This section compares the cooperation model between PCE and VNTM, 628 and the higher-layer signaling trigger model, in terms of VNTM 629 functions, border LSR functions, higher-layer signaling time, and 630 complexity (in terms of number of states and messages). An 631 appropriate model is chosen, taking into all these considerations. 633 Oki et al Expires April 2007 12 634 6.1. VNTM functions: 636 In the cooperation model, VNTM functions are required. In this 637 model, additional overhead communications between PCE and VNTM and 638 between VNTM and a border LSR are required. 640 In the higher-layer signaling trigger model, no VNTM functions are 641 required, and no such communications are required. 643 If VNTM functions are not supported in a multi-layer network, the 644 higher-layer signaling trigger model has to be chosen. 646 The inclusion of VNTM functionality allows better coordination of 647 cross-network LSP tunnels and application of network-wide policy 648 that is not available in the trigger model. 650 6.2. Border LSR functions: 652 In the higher-layer signaling trigger model, a border LSR must have 653 some additional functions. It needs to trigger lower-layer 654 signaling when a higher-layer path message suggests that lower- 655 layer LSP setup is necessary. The triggering signaling is also 656 required in the cooperation case when the VNTM support virtual TE 657 links. Note that, if only the cooperation model is applied, it is 658 required that a PCE knows whether a link is a regular TE link or 659 virtual TE link. 661 If the ERO in the higher-layer Path message uses a mono-layer path 662 or specifies loose hop, a border LSR receiving the Path message 663 MUST obtain a lower-layer route either by consulting PCE or by 664 using its own computation engine. If the ERO in the higher-layer 665 Path message uses multi-layer path, the border LSR MUST judge 666 whether lower-layer signaling is needed. 668 In the cooperation model, no additional function for triggered 669 signaling in border LSRs is required except when virtual TE links 670 are used. Therefore, if these additional functions are not 671 supported in border LSRs, the cooperation model, where a border LSR 672 is controlled by VNTM to set up a lower-layer LSP, has to be chosen. 674 6.3. Complete inter-layer LSP setup time: 676 Complete inter-layer LSP setup time includes inter-layer path 677 computation, signaling, and communication time between PCC and PCE, 678 PCE and VNTM, and VNTM and LSR. In the cooperation model, the 679 additional communication steps are required compared with the 680 higher-layer signaling trigger model. On the other hand, the 681 cooperation model provides better control at the cost of a longer 682 service setup time. 684 Note that, in terms of higher-layer signaling time, in the higher- 685 layer signaling trigger model, the required time from when higher- 686 layer signaling starts to when it is completed, is more than that 688 Oki et al Expires April 2007 13 689 of the cooperation model except when any virtual TE link is 690 included. This is because the former model requires lower-layer 691 signaling to take place during the higher-layer signaling. A 692 higher-layer ingress LSR has to wait for more time until the 693 higher-layer signaling is completed. A higher-layer ingress LSR is 694 required to be tolerant of longer path setup times. 696 An appropriate model is chosen, taking into all of the above 697 considerations. 699 6.4. Network complexity 701 If the higher and lower layer networks have multiple interconnects 702 then optimal path computation for end-to-end LSPs that cross the 703 layer boundaries is non-trivial. The higher layer LSP must be 704 routed to the correct layer border nodes to achieve optimality in 705 both layers. 707 Where the lower layer LSPs are advertised into the higher layer 708 network as TE links, the computation can be resolved in the higher 709 layer network. Care needs to be taken in the allocation of TE 710 metrics (i.e., costs) to the lower layer LSPs as they are 711 advertised as TE links into the higher layer network, and this 712 might be an issue for a VNT Manager component. 714 In the single PCE model an end-to-end path can be found in a single 715 computation because there is full visibility into both layers and 716 all possible paths through all layer interconnects can be 717 considered. 719 Where PCEs cooperate to determine a path, an iterative computation 720 model such as [BRPC] can be used to select an optimal path across 721 layers. 723 When non-cooperating mono-layer PCEs, each of which is in each 724 layer, are used with a triggered LSP model, it is not possible to 725 determine the best border LSRs, and connectivity cannot even be 726 guaranteed. In this case, signaling crankback techniques [CRANK] 727 can be used to eventually achieve connectivity, but optimality is 728 far harder to achieve. In this model, a PCE that is requested to 729 compute a path by an ingress LSR expects a border LSR to setup a 730 lower-layer path triggered by high-layer signaling when there is no 731 TE link between border LSRs. 733 7. Security Considerations 735 Inter-layer traffic engineering with PCE may raise new security 736 issues in both inter-layer path control models. 738 In the cooperation model between PCE and VNTM, when PCE judges a 739 new lower-layer LSP, communications between PCE and VNTM and 740 between VNTM and a border LSR are needed. In this case, there are 741 some security concerns that need to be addressed for these 743 Oki et al Expires April 2007 14 744 communications. These communications should have some security 745 mechanisms to ensure authenticity, privacy and integrity. 747 In the higher-layer signaling trigger model, there are several 748 security concerns. First, PCE may inform PCC, which is located in 749 the higher-layer network, of multi-layer path information that 750 includes an ERO in the lower-layer network, while the PCC may not 751 have TE topology visibility into the lower-layer network. This 752 raises a security concern, where lower-layer hop information is 753 known to transit LSRs supporting a higher-layer LSP. Some security 754 mechanisms to ensure authenticity, privacy and integrity may be 755 used. 757 Security issues may also exist when a single PCE is granted full 758 visibility of TE information that applies to multiple layers. 760 8. Acknowledgment 762 We would like to thank Kohei Shiomoto, Ichiro Inoue, Julien Meuric, 763 Jean-Francois Peltier, Young Lee, and Ina Minei for their useful 764 comments. 766 9. References 768 9.1. Normative Reference 770 [RFC2119] Bradner, S., "Key words for use in RFCs to indicate 771 requirements levels", RFC 2119, March 1997. 773 [RFC3945] Mannie, E., "Generalized Multi-Protocol Label Switching 774 Architecture", RFC 3945, October 2004. 776 [RFC4206] Kompella, K., and Rekhter, Y., "Label Switched Paths 777 (LSP) Hierarchy with Generalized Multi-Protocol Label Switching 778 (GMPLS) Traffic Engineering (TE)", RFC 4206, October 2005. 780 [RFC4208] G. Swallow et al., "Generalized Multiprotocol Label 781 Switching (GMPLS) User-Network Interface (UNI): Resource 782 ReserVation Protocol-Traffic Engineering (RSVP-TE) Support for the 783 Overlay Model", RFC 4208, October 2005. 785 [RFC4655] A. Farrel, JP. Vasseur and J. Ash, "A Path Computation 786 Element (PCE)-Based Architecture", RFC 4655, August 2006. 788 9.2. Informative Reference 790 [RFC4657] J. Ash, J.L Le Roux et al., " Path Communication Element 791 (PCE) Communication Protocol Generic Requirements", RFC 4657, 792 September 2006. 794 [RFC4674] JL Le Roux et al., "Requirements for Path Computation 795 Element (PCE) Discovery", RFC 4674, September 2006. 797 Oki et al Expires April 2007 15 799 [MLN-REQ] K. Shiomoto et al., "Requirements for GMPLS-based multi- 800 region networks (MRN)", draft-ietf-ccamp-gmpls-mln-reqs (work in 801 progress). 803 [PCE-INTER-LAYER-REQ] E. Oki et al., "PCC-PCE Communication 804 Requirements for Inter-Layer Traffic Engineering", draft-ietf-pce- 805 inter-layer-req (work in progress). 807 [PCEP] JP. Vasseur et al, "Path Computation Element (PCE) 808 communication Protocol (PCEP) - Version 1 -" draft-ietf-pce-pcep 809 (work in progress). 811 [BRPC] JP. Vasseur et al., "A Backward Recursive PCE-based 812 Computation (BRPC) procedure to compute shortest inter-domain 813 Traffic Engineering Label Switched Paths", draft-ietf-pce-brpc 814 (work in progress). 816 [CRANK] A. Farrel et al., "Crankback Signaling Extensions for MPLS 817 and GMPLS RSVP-TE", draft-ietf-ccamp-crankback (work in progress). 819 10. Authors' Addresses 821 Eiji Oki 822 NTT 823 3-9-11 Midori-cho, 824 Musashino-shi, Tokyo 180-8585, Japan 825 Email: oki.eiji@lab.ntt.co.jp 827 Jean-Louis Le Roux 828 France Telecom R&D, 829 Av Pierre Marzin, 830 22300 Lannion, France 831 Email: jeanlouis.leroux@orange-ft.com 833 Adrian Farrel 834 Old Dog Consulting 835 Email: adrian@olddog.co.uk 837 11. Intellectual Property Statement 839 The IETF takes no position regarding the validity or scope of any 840 Intellectual Property Rights or other rights that might be claimed 841 to pertain to the implementation or use of the technology described 842 in this document or the extent to which any license under such 843 rights might or might not be available; nor does it represent that 844 it has made any independent effort to identify any such rights. 845 Information on the procedures with respect to rights in RFC 846 documents can be found in BCP 78 and BCP 79. 848 Copies of IPR disclosures made to the IETF Secretariat and any 849 assurances of licenses to be made available, or the result of an 850 attempt made to obtain a general license or permission for the use 851 of such proprietary rights by implementers or users of this 853 Oki et al Expires April 2007 16 854 specification can be obtained from the IETF on-line IPR repository 855 at http://www.ietf.org/ipr. 857 The IETF invites any interested party to bring to its attention any 858 copyrights, patents or patent applications, or other proprietary 859 rights that may cover technology that may be required to implement 860 this standard. Please address the information to the IETF at ietf- 861 ipr@ietf.org. 863 Disclaimer of Validity 865 This document and the information contained herein are provided on 866 an "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE 867 REPRESENTS OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND 868 THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, 869 EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT 870 THE USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR 871 ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A 872 PARTICULAR PURPOSE. 874 Copyright Statement 876 Copyright (C) The Internet Society (2006). This document is 877 subject to the rights, licenses and restrictions contained in BCP 878 78, and except as set forth therein, the authors retain all their 879 rights. 881 Oki et al Expires April 2007 17