idnits 2.17.1 draft-ietf-teas-te-express-path-00.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 (January 28, 2015) is 3347 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- == Outdated reference: A later version (-11) exists of draft-ietf-isis-te-metric-extensions-04 Summary: 0 errors (**), 0 flaws (~~), 2 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 TEAS Working Group A. Atlas 3 Internet-Draft J. Drake 4 Intended status: Informational Juniper Networks 5 Expires: August 1, 2015 S. Giacalone 6 Thomson Reuters 7 D. Ward 8 S. Previdi 9 C. Filsfils 10 Cisco Systems 11 January 28, 2015 13 Performance-based Path Selection for Explicitly Routed LSPs using TE 14 Metric Extensions 15 draft-ietf-teas-te-express-path-00 17 Abstract 19 In certain networks, it is critical to consider network performance 20 criteria when selecting the path for an explicitly routed RSVP-TE 21 LSP. Such performance criteria can include latency, jitter, and loss 22 or other indications such as the conformance to link performance 23 objectives and non-RSVP TE traffic load. This specification uses 24 network performance data, such as is advertised via the OSPF and ISIS 25 TE metric extensions (defined outside the scope of this document) to 26 perform such path selections. 28 Status of This Memo 30 This Internet-Draft is submitted in full conformance with the 31 provisions of BCP 78 and BCP 79. 33 Internet-Drafts are working documents of the Internet Engineering 34 Task Force (IETF). Note that other groups may also distribute 35 working documents as Internet-Drafts. The list of current Internet- 36 Drafts is at http://datatracker.ietf.org/drafts/current/. 38 Internet-Drafts are draft documents valid for a maximum of six months 39 and may be updated, replaced, or obsoleted by other documents at any 40 time. It is inappropriate to use Internet-Drafts as reference 41 material or to cite them other than as "work in progress." 43 This Internet-Draft will expire on August 1, 2015. 45 Copyright Notice 47 Copyright (c) 2015 IETF Trust and the persons identified as the 48 document authors. All rights reserved. 50 This document is subject to BCP 78 and the IETF Trust's Legal 51 Provisions Relating to IETF Documents 52 (http://trustee.ietf.org/license-info) in effect on the date of 53 publication of this document. Please review these documents 54 carefully, as they describe your rights and restrictions with respect 55 to this document. Code Components extracted from this document must 56 include Simplified BSD License text as described in Section 4.e of 57 the Trust Legal Provisions and are provided without warranty as 58 described in the Simplified BSD License. 60 Table of Contents 62 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 63 1.1. Basic Requirements . . . . . . . . . . . . . . . . . . . 4 64 1.2. Oscillation and Stability Considerations . . . . . . . . 4 65 2. Using Performance Data Constraints . . . . . . . . . . . . . 5 66 2.1. End-to-End Constraints . . . . . . . . . . . . . . . . . 5 67 2.2. Link Constraints . . . . . . . . . . . . . . . . . . . . 6 68 2.3. Links out of compliance with Link Performance Objectives 6 69 2.3.1. Use of Anomalous Links for New Paths . . . . . . . . 7 70 2.3.2. Links entering the Anomalous State . . . . . . . . . 7 71 2.3.3. Links leaving the Anomalous State . . . . . . . . . . 8 72 3. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 8 73 4. Security Considerations . . . . . . . . . . . . . . . . . . . 8 74 5. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 8 75 6. References . . . . . . . . . . . . . . . . . . . . . . . . . 8 76 6.1. Normative References . . . . . . . . . . . . . . . . . . 8 77 6.2. Informative References . . . . . . . . . . . . . . . . . 9 78 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 9 80 1. Introduction 82 In certain networks, such as financial information networks, network 83 performance information is becoming as critical to data path 84 selection as other existing metrics. Network performance information 85 can be obtained via either the TE Metric Extensions in OSPF 86 [I-D.ietf-ospf-te-metric-extensions] or ISIS 87 [I-D.ietf-isis-te-metric-extensions] or via a management system. As 88 with other TE information flooded via OSPF or ISIS, the TE metric 89 extensions have a flooding scope limited to the local area or level. 90 This document describes how to use that information for path 91 selection for explicitly routed LSPs signaled via RSVP-TE [RFC3209]. 92 Methods of optimizing path selection for multiple parameters are 93 generally computationally complex. However, there are good 94 heuristics for the delay-constrained lowest-cost (DCLC) computation 95 problem [k-Paths_DCLC] that can be applied to consider both path cost 96 and a maximum delay bound. Some of the network performance 97 information can also be used to prune links from a topology before 98 computing the path. 100 The path selection mechanisms described in this document apply to 101 paths that are fully computed by the head-end of the LSP and then 102 signaled in an ERO where every sub-object is strict. This allows the 103 head-end to consider IGP-distributed performance data without 104 requiring the ability to signal the performance constraints in an 105 object of the RSVP Path message. 107 When considering performance-based data, it is obvious that there are 108 additional contributors to latency beyond just the links. Clearly 109 end-to-end latency is a combination of router latency (e.g. latency 110 from traversing a router without queueing delay), queuing latency, 111 physical link latency and other factors. While traversing a router 112 can cause delay, that router latency can be included in the 113 advertised link delay. As described in 114 [I-D.ietf-ospf-te-metric-extensions] and 115 [I-D.ietf-isis-te-metric-extensions], queuing delay must not be 116 included in the measurements advertised by OSPF or ISIS. 118 Queuing latency is specifically excluded to insure freedom from 119 oscillations and stability issues that have plagued prior attempts to 120 use delay as a routing metric. If application traffic follows a path 121 based upon latency constraints, the same traffic might be in an 122 Expedited Forwarding Per-Hop-Behavior [RFC3246] with minimal queuing 123 delay or another PHB with potentially very substantial per-hop 124 queuing delay. Only traffic which experiences relatively low 125 congestion, such as Expedited Forwarding traffic, will experience 126 delays very close to the sum of the reported link delays. 128 This document does not specify how a router determines what values to 129 advertise by the IGP; it does assume that the constraints specified 130 in [I-D.ietf-ospf-te-metric-extensions] and 131 [I-D.ietf-isis-te-metric-extensions] are followed. Additionally, the 132 end-to-end performance that is computed for an LSP path should be 133 built from the individual link data. Any end-to-end characterization 134 used to determine an LSP's performance compliance should be fully 135 reflected in the Traffic Engineering Database so that a path 136 calculation can also determine whether a path under consideration 137 would be in compliance. 139 1.1. Basic Requirements 141 The following are the requirements that motivate this solution. 143 1. Select a TE tunnel's path based upon a combination of existing 144 constraints as well as on link-latency, packet loss, jitter, link 145 performance objectives conformance, and bandwidth consumed by 146 non-RSVP-TE traffic. 148 2. Ability to define different end-to-end performance requirements 149 for each TE tunnel regardless of common use of resources. 151 3. Ability to periodically verify with the TE LSDB that a TE 152 tunnel's current LSP complies with its configured end-to-end 153 performance requirements. 155 4. Ability to move tunnels, using make-before-break, based upon 156 computed end-to-end performance complying with constraints. 158 5. Ability to move tunnels away from any link that is violating an 159 underlying link performance objective. 161 6. Ability to optionally avoid setting up tunnels using any link 162 that is violating a link performance objective, regardless of 163 whether end-to-end performance would still meet requirements. 165 7. Ability to revert back using make-before-break to the best path 166 after a configurable period. 168 1.2. Oscillation and Stability Considerations 170 Past attempts to use unbounded delay or loss as metric sufferred from 171 severe oscillations. The use of performance based data must be such 172 that undampened oscillations are not possible and stability cannot be 173 impacted. 175 The use of timers is often cited as a cure. Oscillation that is 176 damped by timers is known as "slosh". If advertisement timers are 177 very short relative to the jitter applied to RSVP-TE CSPF timers, 178 then a partial oscillation occurs. If RSVP-TE CSPF timers are short 179 relative to advertisement timers, full oscillation (all traffic 180 moving back and forth) can occur. Even a partial oscillation causes 181 unnecessary reordering which is considered at least minimally 182 disruptive. 184 Delay variation or jitter is affected by even small traffic levels. 185 At even tiny traffic levels, the probability of a queue occupancy of 186 one can produce a measured jitter proportional to or equal to the 187 packet serialization delay. Very low levels of traffic can increase 188 the probability of queue occupancies of two or three packets enough 189 to further increase the measured jitter. Because jitter measurement 190 is extremely sensitive to even very low traffic levels, any use of 191 jitter is likely to oscillate. There may be legitimate use of a 192 jitter measurement in path computation that can be considered free of 193 oscillation. 195 Delay measurements that are not sensitive to traffic loads may be 196 safely used in path computation. Delay measurements made at the link 197 layer or measurements made at a queuing priority higher than any 198 significant traffic (such as DSCP CS7 or CS6 [RFC4594], but not CS2 199 if traffic levels at CS3 and higher or EF and AF can affect the 200 measurement). Making delay measurements at the same priority as the 201 traffic on affected paths is likely to cause oscillations. 203 2. Using Performance Data Constraints 205 2.1. End-to-End Constraints 207 The per-link performance data available in the IGP 208 [I-D.ietf-ospf-te-metric-extensions] 209 [I-D.ietf-isis-te-metric-extensions] includes: unidirectional link 210 delay, unidirectional delay variation, and link loss. Each (or all) 211 of these parameters can be used to create the path-level link-based 212 parameter. 214 It is possible to compute a CSPF where the link latency values are 215 used instead of TE metrics, this results in ignoring the TE metrics 216 and causing LSPs to prefer the lowest-latency paths. In practical 217 scenarios, latency constraints are typically a bound constraint 218 rather than a minimization objective. An end-to-end latency upper 219 bound merely requires that the path computed be no more than that 220 bound and does not require that it be the minimum latency path. The 221 latter is exactly the delay-constrained lowest-cost (DCLC) problem to 222 which good heuristics have been proposed in the literature (e.g. 223 [k-Paths_DCLC]). 225 An end-to-end bound on delay variation can be used similarly as a 226 constraint in the path computation on what links to explore where the 227 path's delay variation is the sum of the used links' delay 228 variations. 230 For link loss, the path loss is not the sum of the used links' 231 losses. Instead, the path loss percentage is 100 - (100 - 232 loss_L1)*(100 - loss_L2)*...*(100 - loss_Ln), where the links along 233 the path are L1 to Ln. The end-to-end link loss bound, computed in 234 this fashion, can also be used as a constraint in the path 235 computation. 237 The heuristic algorithms for DCLC only address one constraint bound 238 but having a CSPF that limits the paths explored (i.e. based on hop- 239 count) can be combined [hop-count_DCLC]. 241 2.2. Link Constraints 243 In addition to selecting paths that conform to a bound on performance 244 data, it is also useful to avoid using links that do not meet a 245 necessary constraint. Naturally, if such a parameter were a known 246 fixed value, then resource attribute flags could be used to express 247 this behavior. However, when the parameter associated with a link 248 may vary dynamically, there is not currently a configuration-time 249 mechanism to enforce such behavior. An example of this is described 250 in Section 2.3, where links may move in and out of conformance for 251 link performance objectives with regards to latency, delay variation, 252 and link loss. 254 When doing path selection for TE tunnels, it has not been possible to 255 know how much actual bandwidth is available that includes the 256 bandwidth used by non-RSVP-TE traffic. In 257 [I-D.ietf-ospf-te-metric-extensions] 258 [I-D.ietf-isis-te-metric-extensions], the Unidirectional Available 259 Bandwidth is advertised as is the Residual Bandwidth. When computing 260 the path for a TE tunnel, only links with at least a minimum amount 261 of Unidirectional Available Bandwidth might be permitted. 263 Similarly, only links whose loss is under a configurable value might 264 be acceptable. For these constraints, each link can be tested 265 against the constraint and only explored in the path computation if 266 the link passes. In essence, a link that fails the constraint test 267 is treated as if it contained a resource attribute in the exclude-any 268 filter. 270 2.3. Links out of compliance with Link Performance Objectives 272 Link conformance to a link performance objective can change as a 273 result of rerouting at lower layers. This could be due to optical 274 regrooming or simply rerouting of a FA-LSP. When this occurs, there 275 are two questions to be asked: 277 a. Should the link be trusted and used for the setup of new LSPs? 279 b. Should LSPs using this link automatically be moved to a secondary 280 path? 282 2.3.1. Use of Anomalous Links for New Paths 284 If the answer to (a) is no for link latency performance objectives, 285 then any link which has the Anomalous bit set in the Unidirectional 286 Link Delay sub-TLV[I-D.ietf-ospf-te-metric-extensions] 287 [I-D.ietf-isis-te-metric-extensions] should be removed from the 288 topology before a path calculation is used to compute a new path. In 289 essence, the link should be treated exactly as if it fails the 290 exclude-any resource attributes filter.[RFC3209]. 292 Similarly, if the answer to (a) is no for link loss performance 293 objectives, then any link which has the Anomalous bit set in the Link 294 Los sub-TLV should be treated as if it fails the exclude-any resource 295 attributes filter. If the answer to (a) is no for link jitter 296 performance objectives, then any link that has the Anomalous bit set 297 in the Unidirectional Delay Variation sub- 298 TLV[I-D.ietf-isis-te-metric-extensions] should be treated as if it 299 fails the exclude-any resource attributes filter. 301 2.3.2. Links entering the Anomalous State 303 When a link enters the Anomalous state with respect to a parameter, 304 this is an indication that LSPs using that link might also no longer 305 be in compliance with their performance bounds. It can also be 306 considered an indication that something is changing that link and so 307 it might no longer be trustworthy to carry performance-critical 308 traffic. Naturally, which performance criteria are important for a 309 particular LSP is dependent upon the LSP's configuration and thus the 310 compliance of a link with respect to a particular link performance 311 objective is indicated per performance criterion. 313 At the ingress of a TE tunnel, a TE tunnel may be configured to be 314 sensitive to the Anomalous state of links in reference to latency, 315 delay variation, and/or loss. Additionally, such a TE tunnel may be 316 configured to either verify continued compliance, to switch 317 immediately to a standby LSP, or to move to a different path. 319 When a sub-TLV is received with the Anomalous bit set when previously 320 it was clear, the list of interested TE tunnels must be scanned. 321 Each such TE tunnel should either have its continued compliance 322 verified, be switched to a hot standby, or do a make-before-break to 323 a secondary path. 325 It is not sufficient to just look at the Anomalous bit in order to 326 determine when TE tunnels must have their compliance verified. When 327 changing to set, the Anomalous bit merely provides a hint that 328 interested TE tunnels should have their continued compliance 329 verified. 331 2.3.3. Links leaving the Anomalous State 333 When a link leaves the Anomalous state with respect to a parameter, 334 this can serve as an indication that those TE tunnels, whose LSPs 335 were changed due to administrative policy when the link entered the 336 Anomalous state, may want to reoptimize to a better path. The hint 337 provided by the Anomalous state change may help optimize when to 338 recompute for a better path. 340 3. IANA Considerations 342 This document includes no request to IANA. 344 4. Security Considerations 346 This document is not currently believed to introduce new security 347 concerns. 349 5. Acknowledgements 351 The authors would like to thank Curtis Villamizar for his extensive 352 detailed comments and suggested text in the Section 1 and 353 Section 1.2. The authors would also like to thank Xiaohu Xu and 354 Sriganesh Kini for their review. 356 6. References 358 6.1. Normative References 360 [I-D.ietf-isis-te-metric-extensions] 361 Previdi, S., Giacalone, S., Ward, D., Drake, J., Atlas, 362 A., Filsfils, C., and W. Wu, "IS-IS Traffic Engineering 363 (TE) Metric Extensions", draft-ietf-isis-te-metric- 364 extensions-04 (work in progress), October 2014. 366 [I-D.ietf-ospf-te-metric-extensions] 367 Giacalone, S., Ward, D., Drake, J., Atlas, A., and S. 368 Previdi, "OSPF Traffic Engineering (TE) Metric 369 Extensions", draft-ietf-ospf-te-metric-extensions-11 (work 370 in progress), January 2015. 372 [RFC3209] Awduche, D., Berger, L., Gan, D., Li, T., Srinivasan, V., 373 and G. Swallow, "RSVP-TE: Extensions to RSVP for LSP 374 Tunnels", RFC 3209, December 2001. 376 6.2. Informative References 378 [RFC3246] Davie, B., Charny, A., Bennet, J., Benson, K., Le Boudec, 379 J., Courtney, W., Davari, S., Firoiu, V., and D. 380 Stiliadis, "An Expedited Forwarding PHB (Per-Hop 381 Behavior)", RFC 3246, March 2002. 383 [RFC4594] Babiarz, J., Chan, K., and F. Baker, "Configuration 384 Guidelines for DiffServ Service Classes", RFC 4594, August 385 2006. 387 [hop-count_DCLC] 388 Agrawal, H., Grah, M., and M. Gregory, "Optimization of 389 QoS Routing", 6th IEEE/AACIS International Conference on 390 Computer and Information Science 2007, 2007, 391 . 394 [k-Paths_DCLC] 395 Jia, Z. and P. Varaiya, "Heuristic methods for delay 396 constrained least cost routing using k-shortest-paths", 397 IEEE Transactions on Automatic Control 51(4), 2006, 398 . 401 Authors' Addresses 403 Alia Atlas 404 Juniper Networks 405 10 Technology Park Drive 406 Westford, MA 01886 407 USA 409 Email: akatlas@juniper.net 411 John Drake 412 Juniper Networks 413 1194 N. Mathilda Ave. 414 Sunnyvale, CA 94089 415 USA 417 Email: jdrake@juniper.net 418 Spencer Giacalone 419 Thomson Reuters 420 195 Broadway 421 New York, NY 10007 422 USA 424 Email: Spencer.giacalone@thomsonreuters.com 426 Dave Ward 427 Cisco Systems 428 170 West Tasman Dr. 429 San Jose, CA 95134 430 USA 432 Email: dward@cisco.com 434 Stefano Previdi 435 Cisco Systems 436 Via Del Serafico 200 437 Rome 00142 438 Italy 440 Email: sprevidi@cisco.com 442 Clarence Filsfils 443 Cisco Systems 444 Brussels 445 Belgium 447 Email: cfilsfil@cisco.com