idnits 2.17.1 draft-wang-teas-pce-native-ip-06.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- -- The document has an IETF Trust Provisions (28 Dec 2009) Section 6.c(ii) Publication Limitation clause. If this document is intended for submission to the IESG for publication, this constitutes an error. 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 : ---------------------------------------------------------------------------- ** There are 2 instances of too long lines in the document, the longest one being 1 character in excess of 72. ** The abstract seems to contain references ([RFC8283], [I-D.draft-wang-teas-ccdr]), which it shouldn't. Please replace those with straight textual mentions of the documents in question. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == Line 97 has weird spacing: '...scribes the ...' == Line 105 has weird spacing: '... 5) Can util...' == Line 133 has weird spacing: '...traffic betwe...' == Line 188 has weird spacing: '... that illus...' == Line 200 has weird spacing: '...routers respe...' == The document doesn't use any RFC 2119 keywords, yet seems to have RFC 2119 boilerplate text. -- The document date (January 19, 2018) is 2282 days in the past. Is this intentional? -- Found something which looks like a code comment -- if you have code sections in the document, please surround them with '' and '' lines. Checking references for intended status: Experimental ---------------------------------------------------------------------------- == Missing Reference: 'RFC2119' is mentioned on line 123, but not defined == Unused Reference: 'RFC4655' is defined on line 378, but no explicit reference was found in the text Summary: 2 errors (**), 0 flaws (~~), 9 warnings (==), 3 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 1 TEAS Working Group A.Wang 2 Internet Draft China Telecom 3 Quintin Zhao 4 Huawei Technologies 5 Boris Khasanov 6 Huawei Technologies 7 HuaiMo Chen 8 Huawei Technologies 9 Penghui Mi 10 Tencent Company 12 Intended status: Experimental Track January 19, 2018 13 Expires: July 18, 2018 15 PCE in Native IP Network 16 draft-wang-teas-pce-native-ip-06.txt 18 Status of this Memo 20 This Internet-Draft is submitted in full conformance with the 21 provisions of BCP 78 and BCP 79. 23 This Internet-Draft is submitted in full conformance with the 24 provisions of BCP 78 and BCP 79. This document may not be modified, 25 and derivative works of it may not be created, and it may not be 26 published except as an Internet-Draft. 28 This Internet-Draft is submitted in full conformance with the 29 provisions of BCP 78 and BCP 79. This document may not be modified, 30 and derivative works of it may not be created, except to publish it 31 as an RFC and to translate it into languages other than English. 33 it for publication as an RFC or to translate it into languages other 34 than English. 36 Internet-Drafts are working documents of the Internet Engineering 37 Task Force (IETF), its areas, and its working groups. Note that 38 other groups may also distribute working documents as Internet- 39 Drafts. 41 Internet-Drafts are draft documents valid for a maximum of six 42 months and may be updated, replaced, or obsoleted by other documents 43 at any time. It is inappropriate to use Internet-Drafts as 44 reference material or to cite them other than as "work in progress." 46 The list of current Internet-Drafts can be accessed at 47 http://www.ietf.org/ietf/1id-abstracts.txt 49 The list of Internet-Draft Shadow Directories can be accessed at 50 http://www.ietf.org/shadow.html 51 This Internet-Draft will expire on July 18, 2018. 53 Copyright Notice 55 Copyright (c) 2017 IETF Trust and the persons identified as the 56 document authors. All rights reserved. 58 This document is subject to BCP 78 and the IETF Trust's Legal 59 Provisions Relating to IETF Documents 60 (http://trustee.ietf.org/license-info) in effect on the date of 61 publication of this document. Please review these documents 62 carefully, as they describe your rights and restrictions with 63 respect to this document. 65 Abstract 67 This document defines the framework for CCDR traffic engineering within 68 Native IP network, using Dual/Multi-BGP session strategy and PCE-based 69 central control architecture. 71 The proposed central mode control framework conforms to the concept 72 that defined in draft [RFC8283]. 74 The scenario and simulation results of CCDR traffic engineering is 75 described in draft [I-D.draft-wang-teas-ccdr] 77 Table of Contents 79 1. Introduction ................................................ 2 80 2. Conventions used in this document............................ 3 81 3. Dual-BGP framework for simple topology. ......................3 82 4. Dual-BGP in large Scale Topology............................. 5 83 5. Multi-BGP for Extended Traffic Differentiation ...............5 84 6. CCDR based framework for Multi-BGP strategy deployment....... 6 85 7. PCEP extension for key parameters delivery................... 7 86 8. CCDR Deployment Consideration................................ 8 87 9. Security Considerations...................................... 9 88 10. IANA Considerations......................................... 9 89 11. Conclusions ................................................ 9 90 12. References ................................................. 9 91 12.1. Normative References................................... 9 92 12.2. Informative References................................ 10 93 13. Acknowledgments ........................................... 10 95 1. Introduction 97 Draft [I-D.draft-wang-teas-ccdr] describes the scenario and 98 simulation results for the CCDR traffic engineering. In summary, the 99 requirements for CCDR traffic engineering in Native IP network are 100 the following: 101 1) No complex MPLS signaling procedure. 102 2) End to End traffic assurance, determined QoS behavior. 103 3) Identical deployment method for intra- and inter- domain. 104 4) No influence to existing router forward behavior. 105 5) Can utilize the power of centrally control(PCE) and 106 flexibility/robustness of distributed control protocol. 107 6) Coping with the differentiation requirements for large amount 108 traffic and prefixes. 109 7) Flexible deployment and automation control. 111 This document defines the framework for CCDR traffic engineering 112 within Native IP network, using Dual/Multi-BGP session strategy and 113 CCDR architecture, to meet the above requirements in dynamical and 114 central control mode. Future PCEP protocol extensions to transfer the 115 key parameters between PCE and the underlying network devices(PCC) 116 are provided in draft [draft-wang-pcep-extension-native-IP] 118 2. Conventions used in this document 120 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 121 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 122 document are to be interpreted as described in RFC 2119 [RFC2119]. 124 3. Dual-BGP framework for simple topology. 126 Dual-BGP framework for simple topology is illustrated in Fig.1, which 127 is comprised by SW1, SW2, R1, R2. There are multiple physical links 128 between R1 and R2. Traffic between IP11 and IP21 is normal traffic, 129 traffic between IP12 and IP22 is priority traffic that should be 130 treated differently. 132 Only Native IGP/BGP protocol is deployed between R1 and R2. The 133 traffic between each address pair may change timely and the 134 corresponding source/destination addresses of the traffic may also 135 change dynamically. 137 The key idea of the Dual-BGP framework for this simple topology is 138 the following: 139 1) Build two BGP sessions between R1 and R2, via the different 140 loopback address lo0, lo1 on these routers. 142 2) Send different prefixes via the two BGP sessions. (For example, 143 IP11/IP21 via the BGP pair 1 and IP12/IP22 via the BGP pair 2). 144 3) Set the explicit peer route on R1 and R2 respectively for BGP next 145 hop of lo0, lo1 to different physical link address between R1 and 146 R2. 148 So, the traffic between the IP11 and IP21, and the traffic between 149 IP12 and IP22 will go through different physical links between R1 and 150 R2, each type of traffic occupy the different dedicated physical 151 links. 153 If there is more traffic between IP12 and IP22 that needs to be 154 assured , one can add more physical links on R1 and R2 to reach the 155 loopback address lo1(also the next hop for BGP Peer pair2). In this 156 cases the prefixes that advertised by two BGP peer need not be 157 changed. 159 If, for example, there is traffic from another address pair that 160 needs to be assured (for example IP13/IP23), but the total volume of 161 assured traffic does not exceed the capacity of the previous 162 appointed physical links, then one need only to advertise the newly 163 added source/destination prefixes via the BGP peer pair2, then the 164 traffic between IP13/IP23 will go through the assigned dedicated 165 physical links as the traffic between IP12/IP22. 167 Such decouple philosophy gives the network operator more flexible 168 control ability on the network traffic, get the determined QoS 169 assurance effect to meet the application's requirement. No complex 170 MPLS signal procedures is introduced, the router need only support 171 native IP protocol. 172 | BGP Peer Pair2 | 173 +------------------+ 174 |lo1 lo1 | 175 | | 176 | BGP Peer Pair1 | 177 +------------------+ 178 IP12 |lo0 lo0 | IP22 179 IP11 | | IP21 180 SW1-------R1-----------------R2-------SW2 181 Links Group 183 Fig.1 Design Philosophy for Dual-BGP Framework 185 4. Dual-BGP in large Scale Topology 187 When the assured traffic spans across one large scale network, as 188 that illustrated in Fig.2, the dual BGP sessions cannot be 189 established hop by hop especially for the iBGP within one AS. For 190 such scenario, we should consider to use the Route Reflector (RR) to 191 achieve the similar Dual-BGP effect, select one router which performs 192 the role of RR (for example R3 in Fig.2), every other edge router 193 will establish two BGP peer sessions with the RR, using their 194 different loopback addresses respectively. The other two steps for 195 traffic differentiation are same as one described in the Dual-BGP 196 simple topology usage case. 198 For the example shown in Fig.2, if we select the R1-R2-R4-R7 as the 199 dedicated path, then we should set the explicit peer routes on these 200 routers respectively, pointing to the BGP next hop (loopback 201 addresses of R1 and R7, which are used to send the prefix of the 202 assured traffic) to the actual address of the physical link 204 +------------R3--------------+ 205 | | 206 SW1-------R1-------R5---------R6-------R7--------SW2 207 | | | | 208 +-------R2---------R4--------+ 210 Fig.2 Dual-BGP Framework for large scale network 212 5. Multi-BGP for Extended Traffic Differentiation 214 In general situation, several additional traffic differentiation 215 criteria exist, including: 216 o Traffic that requires low latency links and is not sensitive to 217 packet loss 218 o Traffic that requires low packet loss but can endure higher latency 219 o Traffic that requires lowest jitter path 220 o Traffic that requires high bandwidth links 222 These different traffic requirements can be summarized in the 223 following table: 225 +----------+-------------+---------------+-----------------+ 226 | Flow No. | Latency | Packet Loss | Jitter | 227 +----------+-------------+---------------+-----------------+ 228 | 1 | Low | Normal | Don't care | 229 +----------+-------------+---------------+-----------------+ 230 | 2 | Normal | Low | Dont't care | 231 +----------+-------------+---------------+-----------------+ 232 | 3 | Normal | Normal | Low | 233 +----------+-------------+---------------+-----------------+ 234 Table 1. Traffic Requirement Criteria 236 For Flow No.1, we can select the shortest distance path to carry the 237 traffic; for Flow No.2, we can select the idle links to form its end 238 to end path; for Flow No.3, we can let all the traffic pass one 239 single path, no ECMP distribution on the parallel links is required. 241 It is difficult and almost impossible to provide an end-to-end (E2E) 242 path with latency, latency variation, packet loss, and bandwidth 243 utilization constraints to meet the above requirements in large scale 244 IP-based network via the traditional distributed routing protocol, 245 but these requirements can be solved using the CCDR architecture 246 since the PCE has the overall network view, can collect real network 247 topology and network performance information about the underlying 248 network, select the appropriate path to meet the various network 249 performance requirements of different traffic type. 251 6. CCDR based framework for Multi-BGP strategy deployment. 253 With the advent of SDN concepts towards pure IP networks, it is 254 possible now to accomplish the central and dynamic control of network 255 traffic according to the application's various requirements. 257 The procedure to implement the dynamic deployment of Multi-BGP 258 strategy is the following: 259 1) PCE gets topology and link utilization information from the 260 underlying network, calculate the appropriate link path upon 261 application's requirements. 262 2) PCE sends the key parameters to edge/RR routers(R1, R7 and R3 in 263 Fig.3) to build multi-BGP peer relations and advertise different 264 prefixes via them. 265 3) PCE sends the route information to the routers (R1,R2,R4,R7 in 266 Fig.3) on forwarding path via PCEP, to build the path to the BGP 267 next-hop of the advertised prefixes. 269 4) If the assured traffic prefixes were changed but the total volume 270 of assured traffic does not exceed the physical capacity of the 271 previous end-to-end path, then PCE needs only change the related 272 information on edge routers (R1,R7 in Fig.3). 273 5) If volume of the assured traffic exceeds the capacity of previous 274 calculated path, PCE must recalculate the appropriate path to 275 accommodate the exceeding traffic via some new end-to-end physical 276 link. After that PCE needs to update on-path routers to build such 277 path hop by hop. 279 +----+ 280 ***********+PCE +************* 281 * +--*-+ * 282 * / * \ * 283 * * * 284 PCEP* *BGP-LS/SNMP *PCEP 285 * * * 286 * * \ * / 287 \ * / * \ */ 288 \*/-----------R3--------------* 289 | | 290 | | 291 SW1-------R1-------R5---------R6-------R7--------SW2 292 | | | | 293 | | | | 294 +-------R2---------R4--------+ 296 Fig.3 PCE based framework for Multi-BGP deployment 298 7. PCEP extension for key parameters delivery. 300 The PCEP protocol needs to be extended to transfer the following key 301 parameters: 302 1) BGP peer address and advertised prefixes. 303 2) Explicit route information to BGP next hop of advertised prefixes. 305 Once the router receives such information, it should establish the 306 BGP session with the peer appointed in the PCEP message, advertise 307 the prefixes that contained in the corresponding PCEP message, and 308 build the end to end dedicated path hop by hop. Details of 309 communications between PCEP and BGP subsystems in router's control 310 plane are out of scope of this draft and will be described in 311 separate draft.[draft-wang-pce-extension for native IP] 313 The reason why we selected PCEP as the southbound protocol instead of 314 OpenFlow, is that PCEP is suitable for the changes in control plane 315 of the network devices, there OpenFlow dramatically changes the 316 forwarding plane. We also think that the level of centralization that 317 requires by OpenFlow is hardly achievable in many today's SP networks 318 so hybrid BGP+PCEP approach looks much more interesting. 320 8. CCDR Deployment Consideration 322 CCDR framework requires the parallel work of 2 subsystems in router's 323 control plane: PCE (PCEP) and BGP as well as coordination between 324 them, so it might require additional planning work before deployment. 326 8.1 Scalability 328 In CCDR framework, PCE needs only to influence the edge routers for 329 the prefixes differentiation via the multi-BGP deployment. The route 330 information for these prefixes within the on-path routers were 331 distributed via the traditional BGP protocol. Unlike the solution 332 from BGP Flowspec, the on-path router need only keep the specific 333 policy routes to the BGP next-hop of the differentiate prefixes, not 334 the specific routes to the prefixes themselves. This can lessen the 335 burden from the table size of policy based routes for the on-path 336 routers, and has more scalability when comparing with the solution 337 from BGP flowspec or Openflow. 339 8.2 High Availability 341 CCDR framework is based on the traditional distributed IP protocol. 342 If the PCE failed, the forwarding plane will not be impacted, as the 343 BGP session between all devices will not flap, and the forwarding 344 table will remain the same. If one node on the optimal path is failed, 345 the assurance traffic will fall over to the best-effort forwarding 346 path. One can even design several assurance paths to load balance/hot 347 standby the assurance traffic to meet the path failure situation, as 348 done in MPLS FRR. 349 From PCE/SDN-controller HA side we will rely on existing HA solutions 350 of SDN controllers such as clustering. 352 8.3 Incremental deployment 354 Not every router within the network support will support the PCEP 355 extension that defined in [draft-wang-pce-extension-native-IP] 356 simultaneously. For such situations, router on the edge of sub domain 357 can be upgraded first, and then the traffic can be assured between 358 different sub domains. Within each sub domain, the traffic will be 359 forwarded along the best-effort path. Service provider can 360 selectively upgrade the routers on each sub-domain in sequence. 362 9. Security Considerations 364 TBD 366 10. IANA Considerations 368 TBD 370 11. Conclusions 372 TBD 374 12. References 376 12.1. Normative References 378 [RFC4655] Farrel, A., Vasseur, J.-P., and J. Ash, "A Path 380 Computation Element (PCE)-Based Architecture", RFC 382 4655, August 2006,. 384 [RFC5440]Vasseur, JP., Ed., and JL. Le Roux, Ed., "Path 386 Computation Element (PCE) Communication Protocol 388 (PCEP)", RFC 5440, March 2009, 390 . 392 [RFC8283] A.Farrel, Q.Zhao et al.," An Architecture for Use of PCE 393 and the PCE Communication Protocol (PCEP) in a Network with Central 394 Control", [RFC8283], December 2017 396 12.2. Informative References 398 [I-D.draft-wang-teas-ccdr] 400 A.Wang, X.Huang et al. "CCDR Scenario, Simulation and Suggestion" 402 https://datatracker.ietf.org/doc/draft-wang-teas-ccdr/ 404 [I-D. draft-ietf-teas-pcecc-use-cases] 406 Quintin Zhao, Robin Li, Boris Khasanov et al. "The Use Cases for 407 Using PCE as the Central Controller(PCECC) of LSPs 409 https://tools.ietf.org/html/draft-ietf-teas-pcecc-use-cases-00 411 March,2017 413 [draft-wang-pcep-extension for native IP] 415 Aijun Wang, Boris Khasanov et al. "PCEP Extension for Native IP 416 Network" https://datatracker.ietf.org/doc/draft-wang-pce-extension- 417 native-ip/ 419 13. Acknowledgments 421 The authors would like to thank George Swallow, Xia Chen, Jeff 422 Tantsura,Scharf Michael,Daniele Ceccarelli and Dhruv Dhody for their 423 valuable comments and suggestions. 425 The authors would also like to thank Lou Berger, Adrian Farrel, 426 Vishnu Pavan Beeram, Deborah Brungard and King Daniel for their 427 suggestions to put forward this draft. 429 Authors' Addresses 431 Aijun Wang 432 China Telecom 433 Beiqijia Town, Changping District 434 Beijing,China 436 Email: wangaj.bri@chinatelecom.cn 437 Quintin Zhao 438 Huawei Technologies 439 125 Nagog Technology Park 440 Acton, MA 01719 441 USA 443 EMail: quintin.zhao@huawei.com 445 Boris Khasanov 446 Huawei Technologies 447 Moskovskiy Prospekt 97A 448 St.Petersburg 196084 449 Russia 451 EMail: khasanov.boris@huawei.com 453 Huaimo Chen 454 Huawei Technologies 455 Boston,MA, 456 USA 458 Email huaimo.chen@huawei.com 460 Penghui Mi 461 Tencent 462 Tencent Building, Kejizhongyi Avenue, 463 Hi-techPark, Nanshan District,Shenzhen 518057, P.R.China 465 Email kevinmi@tencent.com 467 Raghavendra Mallya 468 Juniper Networks 469 1133 Innovation Way 470 Sunnyvale, California 94089 USA 472 Email: rmallya@juniper.net 474 Shaofu Peng 475 ZTE Corporation 476 No.68 Zijinghua Road,Yuhuatai District 477 Nanjing 210012 478 China 480 Email: peng.shaofu@zte.com.cn