idnits 2.17.1 draft-ietf-teas-pce-native-ip-07.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (June 1, 2020) is 1419 days in the past. Is this intentional? Checking references for intended status: Experimental ---------------------------------------------------------------------------- == Unused Reference: 'RFC2119' is defined on line 463, but no explicit reference was found in the text == Unused Reference: 'RFC4456' is defined on line 468, but no explicit reference was found in the text ** Obsolete normative reference: RFC 7752 (Obsoleted by RFC 9552) == Outdated reference: A later version (-30) exists of draft-ietf-pce-pcep-extension-native-ip-05 Summary: 1 error (**), 0 flaws (~~), 4 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 TEAS Working Group A. Wang 3 Internet-Draft China Telecom 4 Intended status: Experimental B. Khasanov 5 Expires: December 3, 2020 Huawei Technologies 6 Q. Zhao 7 Etheric Networks 8 H. Chen 9 Futurewei 10 June 1, 2020 12 PCE in Native IP Network 13 draft-ietf-teas-pce-native-ip-07 15 Abstract 17 This document defines the framework for traffic engineering within 18 native IP network, using multiple BGP sessions strategy and PCE 19 -based central control architecture. 21 Status of This Memo 23 This Internet-Draft is submitted in full conformance with the 24 provisions of BCP 78 and BCP 79. 26 Internet-Drafts are working documents of the Internet Engineering 27 Task Force (IETF). Note that other groups may also distribute 28 working documents as Internet-Drafts. The list of current Internet- 29 Drafts is at https://datatracker.ietf.org/drafts/current/. 31 Internet-Drafts are draft documents valid for a maximum of six months 32 and may be updated, replaced, or obsoleted by other documents at any 33 time. It is inappropriate to use Internet-Drafts as reference 34 material or to cite them other than as "work in progress." 36 This Internet-Draft will expire on December 3, 2020. 38 Copyright Notice 40 Copyright (c) 2020 IETF Trust and the persons identified as the 41 document authors. All rights reserved. 43 This document is subject to BCP 78 and the IETF Trust's Legal 44 Provisions Relating to IETF Documents 45 (https://trustee.ietf.org/license-info) in effect on the date of 46 publication of this document. Please review these documents 47 carefully, as they describe your rights and restrictions with respect 48 to this document. Code Components extracted from this document must 49 include Simplified BSD License text as described in Section 4.e of 50 the Trust Legal Provisions and are provided without warranty as 51 described in the Simplified BSD License. 53 Table of Contents 55 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 56 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 3 57 3. CCDR Framework in Simple Topology . . . . . . . . . . . . . . 3 58 4. CCDR Framework in Large Scale Topology . . . . . . . . . . . 5 59 5. CCDR Multiple BGP Sessions Strategy . . . . . . . . . . . . . 6 60 6. PCEP Extension for Key Parameters Delivery . . . . . . . . . 8 61 7. Deployment Consideration . . . . . . . . . . . . . . . . . . 9 62 7.1. Scalability . . . . . . . . . . . . . . . . . . . . . . . 9 63 7.2. High Availability . . . . . . . . . . . . . . . . . . . . 9 64 7.3. Incremental deployment . . . . . . . . . . . . . . . . . 10 65 8. Security Considerations . . . . . . . . . . . . . . . . . . . 10 66 9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 10 67 10. Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . 10 68 11. References . . . . . . . . . . . . . . . . . . . . . . . . . 11 69 11.1. Normative References . . . . . . . . . . . . . . . . . . 11 70 11.2. Informative References . . . . . . . . . . . . . . . . . 12 71 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 12 73 1. Introduction 75 [RFC8735] describes the scenarios and simulation results for traffic 76 engineering in native IP network. To meet the requirements of 77 various scenarios, the solution for traffic engineering in native IP 78 network should have the following criteria: 80 o No complex signaling procedures among network nodes like MPLS-TE. 82 o End to End traffic assurance, determined QoS behavior. 84 o Same deployment method for intra-domain and inter-domain. 86 o No upgrade to forwarding behavior of the router. 88 o Support native IPv4 and IPv6 traffic in the same solution. 90 o Can exploit the power of centrally control and flexibility/ 91 robustness of distributed control protocol. 93 o Coping with the differentiation requirements for large amount 94 traffic and prefixes. 96 o Flexible deployment and automation control. 98 This document defines the framework for traffic engineering within 99 native IP network, using multiple BGP session strategy, to meet the 100 above requirements in dynamical and centrally control mode. The 101 framework is referred as Central Control Dynamic Routing (CCDR) 102 framework. It depends on the central control (PCE) element to 103 compute the optimal path for selected traffic, and utilizes the 104 dynamic routing behavior of traditional IGP/BGP protocols to forward 105 such traffic. 107 The control messages between PCE and underlying network node are 108 transmitted via Path Computation Element Communications Protocol 109 (PCEP) protocol. The related PCEP extensions are provided in draft 110 [I-D.ietf-pce-pcep-extension-native-ip]. 112 2. Terminology 114 This document uses the following terms defined in [RFC5440]: PCE, 115 PCEP 117 The following terms are used in this document: 119 o CCDR: Central Control Dynamic Routing 121 o E2E: End to End 123 o ECMP: Equal Cost Multi Path 125 o RR: Route Reflector 127 o SDN: Software Defined Network 129 3. CCDR Framework in Simple Topology 131 Figure 1 illustrates the CCDR framework for traffic engineering in 132 simple topology. The topology is comprised by four devices which are 133 SW1, SW2, R1, R2. There are multiple physical links between R1 and 134 R2. Traffic between prefix PF11(on SW1) and prefix PF21(on SW2) is 135 normal traffic, traffic between prefix PF12(on SW1) and prefix 136 PF22(on SW2) is priority traffic that should be treated differently. 138 In Intra-AS scenario, IGP and BGP are deployed between R1 and R2. In 139 inter-AS scenario, only native BGP protocol is deployed. The traffic 140 between each address pair may change in real time and the 141 corresponding source/destination addresses of the traffic may also 142 change dynamically. 144 The key ideas of the CCDR framework for this simple topology are the 145 followings: 147 o Build two BGP sessions between R1 and R2, via the different 148 loopback addresses on these routers. 150 o Send different prefixes via the established BGP sessions. For 151 example, PF11/PF21 via the BGP session 1 and PF12/PF22 via the BGP 152 session 2. 154 o Set the explicit peer route on R1 and R2 respectively for BGP next 155 hop to different physical link addresses between R1 and R2. Such 156 explicit peer route can be set in the format of static route to 157 BGP peer address, which is different from the route learned from 158 the IGP protocol. 160 After the above actions, the traffic between the PF11 and PF21, and 161 the traffic between PF12 and PF22 will go through different physical 162 links between R1 and R2, each set of traffic pass through different 163 dedicated physical links. 165 If there is more traffic between PF12 and PF22 that needs to be 166 assured , one can add more physical links between R1 and R2 to reach 167 the the next hop for BGP session 2. In this cases the prefixes that 168 advertised by the BGP peers need not be changed. 170 If, for example, there is traffic from another address pair that 171 needs to be assured (for example prefix PF13/PF23), and the total 172 volume of assured traffic does not exceed the capacity of the 173 previously provisioned physical links, one need only to advertise the 174 newly added source/destination prefixes via the BGP session 2. The 175 traffic between PF13/PF23 will go through the assigned dedicated 176 physical links as the traffic between PF12/PF22. 178 Such decouple philosophy gives network operator flexible control 179 capability on the network traffic, achieve the determined QoS 180 assurance effect to meet the application's requirement. No complex 181 signaling procedures like MPLS are introduced, the router needs only 182 support native IP and multiple BGP sessions setup via different 183 loopback addresses. 185 +-----+ 186 +----------+ PCE +--------+ 187 | +-----+ | 188 | | 189 | BGP Session 1(lo11/lo21)| 190 +-------------------------+ 191 | | 192 | BGP Session 2(lo12/lo22)| 193 +-------------------------+ 194 PF12 | | PF22 195 PF11 | | PF21 196 +---+ +-----+-----+ +-----+-----+ +---+ 197 |SW1+---------+(lo11/lo12)+-------------+(lo21/lo22)+--------------+SW2| 198 +---+ | R1 +-------------+ R2 | +---+ 199 +-----------+ +-----------+ 201 Figure 1: CCDR framework in simple topology 203 4. CCDR Framework in Large Scale Topology 205 When the assured traffic spans across the large scale network, as 206 that illustrated in Figure 2, the multiple BGP sessions cannot be 207 established hop by hop, especially for the iBGP within one AS. 209 For such scenario, we should consider to use the Route Reflector (RR) 210 [RFC4456]to achieve the similar effect. Every edge router will 211 establish two BGP sessions with the RR via different loopback 212 addresses respectively. The other steps for traffic differentiation 213 are same as that described in the CCDR framework for simple topology. 215 As shown in Figure 2, if we select R3 as the RR, every edge router(R1 216 and R7 in this example) will build two BGP session with the RR. If 217 the PCE selects the dedicated path as R1-R2-R4-R7, then the operator 218 should set the explicit peer routes via PCEP protocol on these 219 routers respectively, pointing to the BGP next hop (loopback 220 addresses of R1 and R7, which are used to send the prefix of the 221 assured traffic) to the selected forwarding address. 223 +-----+ 224 +----------------+ PCE +------------------+ 225 | +--+--+ | 226 | | | 227 | | | 228 | ++-+ | 229 +------------------+R3+-------------------+ 230 PF12 | +--+ | PF22 231 PF11 | | PF21 232 +---+ ++-+ +--+ +--+ +-++ +---+ 233 |SW1+-------+R1+----------+R5+----------+R6+---------+R7+--------+SW2| 234 +---+ ++-+ +--+ +--+ +-++ +---+ 235 | | 236 | | 237 | +--+ +--+ | 238 +------------+R2+----------+R4+-----------+ 239 +--+ +--+ 240 Figure 2: CCDR framework in large scale network 242 5. CCDR Multiple BGP Sessions Strategy 244 In general situation, different applications may require different 245 QoS criteria, which may include: 247 o Traffic that requires low latency and is not sensitive to packet 248 loss. 250 o Traffic that requires low packet loss and can endure higher 251 latency. 253 o Traffic that requires low jitter. 255 These different traffic requirements can be summarized in the 256 following table: 258 +----------------+-------------+---------------+-----------------+ 259 | Prefix Set No. | Latency | Packet Loss | Jitter | 260 +----------------+-------------+---------------+-----------------+ 261 | 1 | Low | Normal | Don't care | 262 +----------------+-------------+---------------+-----------------+ 263 | 2 | Normal | Low | Dont't care | 264 +----------------+-------------+---------------+-----------------+ 265 | 3 | Normal | Normal | Low | 266 +----------------+-------------+---------------+-----------------+ 267 Table 1. Traffic Requirement Criteria 269 For Prefix Set No.1, we can select the shortest distance path to 270 carry the traffic; for Prefix Set No.2, we can select the path that 271 is comprised by under loading links from end to end; For Prefix Set 272 No.3, we can let all assured traffic pass the determined single path, 273 no Equal Cost Multipath (ECMP) distribution on the parallel links is 274 desired. 276 It is almost impossible to provide an End-to-End (E2E) path with 277 latency, jitter, packet loss constraints to meet the above 278 requirements in large scale IP-based network via the distributed 279 routing protocol, but these requirements can be solved with the 280 assistance of PCE, as that described in [RFC4655] and [RFC8283] 281 because the PCE has the overall network view, can collect real 282 network topology and network performance information about the 283 underlying network, select the appropriate path to meet various 284 network performance requirements of different traffics. 286 The framework to implement the CCDR Multiple BGP sessions strategy 287 are the followings. Here PCE is the main component of the Software 288 Definition Network (SDN) controller and is responsible for optimal 289 path computation for priority traffic. 291 o SDN controller gets topology via BGP-LS[RFC7752] and link 292 utilization information via existing Network Monitor System (NMS) 293 from the underlying network. 295 o PCE calculates the appropriate path upon application's 296 requirements, sends the key parameters to edge/RR routers(R1, R7 297 and R3 in Fig.3) to establish multiple BGP sessions and advertises 298 different prefixes via them. The loopback addresses used for BGP 299 sessions should be planned in advance and distributed in the 300 domain. 302 o PCE sends the route information to the routers (R1,R2,R4,R7 in 303 Fig.3) on forwarding path via PCEP 304 [I-D.ietf-pce-pcep-extension-native-ip], to build the path to the 305 BGP next-hop of the advertised prefixes. 307 o If the assured traffic prefixes were changed but the total volume 308 of assured traffic does not exceed the physical capacity of the 309 previous E2E path, PCE needs only change the prefixed advertised 310 via the edge routers (R1,R7 in Fig.3). 312 o If the volume of assured traffic exceeds the capacity of previous 313 calculated path, PCE can recalculate and add the appropriate paths 314 to accommodate the exceeding traffic. After that, PCE needs to 315 update on-path routers to build the forwarding path hop by hop. 317 +------------+ 318 | Application| 319 +------+-----+ 320 | 321 +--------+---------+ 322 +----------+SDN Controller/PCE+-----------+ 323 | +--------^---------+ | 324 | | | 325 | | | 326 PCEP | BGP-LS|PCEP | PCEP 327 | | | 328 | +v-+ | 329 +------------------+R3+-------------------+ 330 PF12 | +--+ | PF22 331 PF11 | | PF21 332 +---+ +v-+ +--+ +--+ +-v+ +---+ 333 |SW1+-------+R1+----------+R5+----------+R6+---------+R7+--------+SW2| 334 +---+ ++-+ +--+ +--+ +-++ +---+ 335 | | 336 | | 337 | +--+ +--+ | 338 +------------+R2+----------+R4+-----------+ 340 Figure 3: CCDR framework for Multi-BGP deployment 342 6. PCEP Extension for Key Parameters Delivery 344 The PCEP protocol needs to be extended to transfer the following key 345 parameters: 347 o Peer addresses pair that is used to build the BGP session 349 o Advertised prefixes and their associated BGP session. 351 o Explicit route information to BGP next hop of advertised prefixes. 353 Once the router receives such information, it should establish the 354 BGP session with the peer appointed in the PCEP message, advertise 355 the prefixes that contained in the corresponding PCEP message, and 356 build the end to end dedicated path hop by hop. 358 The explicit route created by PCE has the higher priority than the 359 route information created by other protocols, including the route 360 manually configured. 362 All above dynamically created states (BGP sessions, Prefix advertised 363 prefix, Explict route) will be cleared once the connection between 364 the PCE and network devices is interrupted. 366 Details of communications between PCEP and BGP subsystems in router's 367 control plane are out of scope of this draft and will be described in 368 separate draft [I-D.ietf-pce-pcep-extension-native-ip] . 370 The reason that we select PCEP as the southbound protocol instead of 371 OpenFlow, is that PCEP is suitable for the changes in control plane 372 of the network devices, while OpenFlow dramatically changes the 373 forwarding plane. We also think that the level of centralization 374 that required by OpenFlow is hardly achievable in SP networks so 375 hybrid BGP+PCEP approach looks much more interesting. 377 7. Deployment Consideration 379 7.1. Scalability 381 In CCDR framework, PCE needs only influence the edge routers for the 382 prefixes advertisement via the multiple BGP sessions deployment. The 383 route information for these prefixes within the on-path routers were 384 distributed via the BGP protocol. 386 For multiple domain deployment, the PCE need only control the edge 387 router to build multiple eBGP sessions, all other procedures are the 388 same that in one domain. 390 Unlike the solution from BGP Flowspec, the on-path router need only 391 keep the specific policy routes to the BGP next-hop of the 392 differentiate prefixes, not the specific routes to the prefixes 393 themselves. This can lessen the burden from the table size of policy 394 based routes for the on-path routers, and has more expandability when 395 comparing with the solution from BGP flowspec or Openflow. For 396 example, if we want to differentiate 1000 prefixes from the normal 397 traffic, CCDR needs only one explicit peer route in every on-path 398 router, but the BGP flowspec or Openflow needs 1000 policy routes on 399 them. 401 7.2. High Availability 403 The CCDR framework is based on the distributed IP protocol. If the 404 PCE failed, the forwarding plane will not be impacted, as the BGP 405 session between all devices will not flap, and the forwarding table 406 will remain unchanged. 408 If one node on the optimal path is failed, the priority traffic will 409 fall over to the best-effort forwarding path. One can even design 410 several assurance paths to load balance/hot-standby the priority 411 traffic to meet the path failure situation. 413 For high availability of PCE/SDN-controller, operator should rely on 414 existing HA solutions for SDN controller, such as clustering 415 technology and deployment. 417 7.3. Incremental deployment 419 Not every router within the network will support the PCEP extension 420 that defined in [I-D.ietf-pce-pcep-extension-native-ip] 421 simultaneously. 423 For such situations, router on the edge of domain can be upgraded 424 first, and then the traffic can be assured between different domains. 425 Within each domain, the traffic will be forwarded along the best- 426 effort path. Service provider can selectively upgrade the routers on 427 each domain in sequence. 429 8. Security Considerations 431 A PCE assures calculations of E2E path upon the status of network 432 condition and the service requirements in real time. 434 The PCE need consider the explicit route deployment order (for 435 example, from tail router to head router) to eliminate the possible 436 transient traffic loop. 438 CCDR framework described in this draft puts more requirements on the 439 function of PCE and its communication with the underlay devices. 440 Service provider should consider more on the protection of PCE and 441 their communication with the underlay devices, which is described in 442 document [RFC5440] and [RFC8253] 444 CCDR framework does not require the change of forward behavior on the 445 underlay devices, then there will no additional security impact on 446 the devices. 448 9. IANA Considerations 450 This document does not require any IANA actions. 452 10. Acknowledgement 454 The author would like to thank Deborah Brungard, Adrian Farrel, 455 Vishnu Beeram, Lou Berger, Dhruv Dhody, Raghavendra Mallya , Mike 456 Koldychev, Haomian Zheng, Penghui Mi, Shaofu Peng and Jessica Chen 457 for their supports and comments on this draft. 459 11. References 461 11.1. Normative References 463 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 464 Requirement Levels", BCP 14, RFC 2119, 465 DOI 10.17487/RFC2119, March 1997, 466 . 468 [RFC4456] Bates, T., Chen, E., and R. Chandra, "BGP Route 469 Reflection: An Alternative to Full Mesh Internal BGP 470 (IBGP)", RFC 4456, DOI 10.17487/RFC4456, April 2006, 471 . 473 [RFC4655] Farrel, A., Vasseur, J., and J. Ash, "A Path Computation 474 Element (PCE)-Based Architecture", RFC 4655, 475 DOI 10.17487/RFC4655, August 2006, 476 . 478 [RFC5440] Vasseur, JP., Ed. and JL. Le Roux, Ed., "Path Computation 479 Element (PCE) Communication Protocol (PCEP)", RFC 5440, 480 DOI 10.17487/RFC5440, March 2009, 481 . 483 [RFC7752] Gredler, H., Ed., Medved, J., Previdi, S., Farrel, A., and 484 S. Ray, "North-Bound Distribution of Link-State and 485 Traffic Engineering (TE) Information Using BGP", RFC 7752, 486 DOI 10.17487/RFC7752, March 2016, 487 . 489 [RFC8253] Lopez, D., Gonzalez de Dios, O., Wu, Q., and D. Dhody, 490 "PCEPS: Usage of TLS to Provide a Secure Transport for the 491 Path Computation Element Communication Protocol (PCEP)", 492 RFC 8253, DOI 10.17487/RFC8253, October 2017, 493 . 495 [RFC8283] Farrel, A., Ed., Zhao, Q., Ed., Li, Z., and C. Zhou, "An 496 Architecture for Use of PCE and the PCE Communication 497 Protocol (PCEP) in a Network with Central Control", 498 RFC 8283, DOI 10.17487/RFC8283, December 2017, 499 . 501 [RFC8735] Wang, A., Huang, X., Kou, C., Li, Z., and P. Mi, 502 "Scenarios and Simulation Results of PCE in a Native IP 503 Network", RFC 8735, DOI 10.17487/RFC8735, February 2020, 504 . 506 11.2. Informative References 508 [I-D.ietf-pce-pcep-extension-native-ip] 509 Wang, A., Khasanov, B., Fang, S., and C. Zhu, "PCEP 510 Extension for Native IP Network", draft-ietf-pce-pcep- 511 extension-native-ip-05 (work in progress), February 2020. 513 Authors' Addresses 515 Aijun Wang 516 China Telecom 517 Beiqijia Town, Changping District 518 Beijing 102209 519 China 521 Email: wangaj3@chinatelecom.cn 523 Boris Khasanov 524 Huawei Technologies 525 Moskovskiy Prospekt 97A 526 St.Petersburg 196084 527 Russia 529 Email: khasanov.boris@huawei.com 531 Quintin Zhao 532 Etheric Networks 533 1009 S CLAREMONT ST 534 SAN MATEO, CA 94402 535 USA 537 Email: qzhao@ethericnetworks.com 539 Huaimo Chen 540 Futurewei 541 Boston, MA 542 USA 544 Email: huaimo.chen@futurewei.com