idnits 2.17.1 draft-kishjac-bmwg-evpntest-08.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: ---------------------------------------------------------------------------- == The page length should not exceed 58 lines per page, but there was 2 longer pages, the longest (page 4) being 75 lines Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- ** There are 52 instances of weird spacing in the document. Is it really formatted ragged-right, rather than justified? ** There are 34 instances of too long lines in the document, the longest one being 54 characters in excess of 72. ** There are 7 instances of lines with control characters in the document. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == Line 29 has weird spacing: '... months and ...' == Line 30 has weird spacing: '... at any time...' == Line 31 has weird spacing: '...ference mate...' == Line 128 has weird spacing: '...in this docu...' == Line 352 has weird spacing: '...s taken by c...' == (47 more instances...) == The document doesn't use any RFC 2119 keywords, yet seems to have RFC 2119 boilerplate text. -- Couldn't find a document date in the document -- date freshness check skipped. Checking references for intended status: Full Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Unused Reference: 'RFC2544' is defined on line 997, but no explicit reference was found in the text == Unused Reference: 'RFC2889' is defined on line 1002, but no explicit reference was found in the text == Unused Reference: 'RFC7432' is defined on line 1007, but no explicit reference was found in the text == Unused Reference: 'RFC7623' is defined on line 1013, but no explicit reference was found in the text ** Downref: Normative reference to an Informational RFC: RFC 2544 ** Downref: Normative reference to an Informational RFC: RFC 2889 Summary: 5 errors (**), 0 flaws (~~), 13 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 INTERNET-DRAFT Kishore Tiruveedhula 3 Intended Status:Standard Sudhin Jacob 4 Expires: March 7,2018 Juniper Networks 5 October 6,2017 7 Benchmarking Methodology for EVPN and PBB-EVPN 8 draft-kishjac-bmwg-evpntest-08 10 Abstract 12 This document defines the methodologies for benchmarking performance 13 of EVPN and PBB-EVPN.EVPN is defined in RFC 7432.It is being deployed 14 in provider network.This document provides the benchmarking 15 methodologies for EVPN/PBB-EVPN convergence,data plane,control plane 16 learning of mac. 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 Internet-Drafts are working documents of the Internet Engineering 24 Task Force (IETF). Note that other groups may also distribute 25 working documents as Internet-Drafts. The list of current Internet- 26 Drafts is at http://datatracker.ietf.org/drafts/current/. 28 Internet-Drafts are draft documents valid for a maximum of six 29 months and may be updated, replaced, or obsoleted by other 30 documents at any time. It is inappropriate to use Internet-Drafts 31 as reference material or to cite them other than as "work in 32 progress." 34 This Internet-Draft will expire on March 7,2018. 36 Copyright and License Notice 38 Copyright (c) 2016 IETF Trust and the persons identified as the 39 document authors. All rights reserved. 41 This document is subject to BCP 78 and the IETF Trust's Legal 42 Provisions Relating to IETF Documents 43 (http://trustee.ietf.org/license-info) in effect on the date of 44 publication of this document. Please review these documents 45 carefully, as they describe your rights and restrictions with respect 46 to this document. Code Components extracted from this document must 47 include Simplified BSD License text as described in Section 4.e of 48 the Trust Legal Provisions and are provided without warranty as 49 described in the Simplified BSD License. 51 INTERNET DRAFT 53 to this document. Code Components extracted from this document must 54 include Simplified BSD License text as described in Section 4.e of 55 the Trust Legal Provisions and are provided without warranty as 56 described in the Simplified BSD License. 58 Table of Contents 60 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 4 61 1.1 Terminology . . . . . . . . . . . . . . . . . . . . . . . . 4 62 2.Test Topology . . . . . . . . . . . . . . . . . . . . . . . . . 4 63 2.1 Terminologies . . . . . . . . . . . . . . . . . . . . . . . 5 64 3. Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 65 3.1 PBB-EVPN Network Setup . . . . . . . . . . . . . . . . . . 5 66 4. Test Procedure . . . . . . . . . . . . . . . . . . . . . . . . 6 67 5 Test Cases . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 68 5.1.1 To check the time taken to learn the mac address in 69 DUT . . . . . . . . . . . . . . . . . . . . . . . . . . 7 70 5.1.2. To check the time taken to learn X macs from remote 71 peer by DUT . . . . . . . . . . . . . . . . . . . . . 9 72 5.1.3. To check the time taken to flush the local entry due 73 to CE link failure . . . . . . . . . . . . . . . . . . 10 74 5.1.4. To check the time taken by DUT to flush X routes 75 learned from remote PE after R1 traffic generator link failure . 11 76 5.1.5. To measure the mac ageing time. . . . . . . . . . . . 12 77 5.1.6. To check the time taken by DUT to age X routes 78 learned from remote PE after stopping the traffic at remote PE. . 14 79 5.1.7. To check the time taken by DUT to learn X routes 80 from local and X from remote and measure the time of flood from DUT. . 15 81 5.1.8. To measure the time taken to elect a new DF by 82 adding a a MHPE. . . . . . . . . . . . . . . . . . . . 16 83 5.2 High Availability . . . . . . . . . . . . . . . . . . . 18 84 5.2.1 To check the whether there is traffic loss due to 85 routing engine failover for redundancy test. . . . . . 18 86 5.3 ARP/ND Scaling . . . . . . . . . . . . . . . . . . . . . . 19 88 6. Scale . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 89 6.1. To Scale the DUT to N EVI and clear bgp in DUT with out 90 traffic. . . . . . . . . . . . . . . . . . . . . . . . . . 20 91 6.2. To Scale the DUT to N EVI and clear bgp in DUT with 92 traffic.Measure the convergence time . . . . . . . . . . . 22 93 7. Soak Test . . . . . . . . . . . . . . . . . . . . . . . . . . 23 94 7.1. To Scale the DUT to N EVI in DUT with traffic and run 95 the set up for 24hrs . . . . . . . . . . . . . . . . . . . 23 96 8. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 25 97 9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 25 98 10. Security Considerations . . . . . . . . . . . . . . . . . . . 25 100 INTERNET DRAFT 102 11 References . . . . . . . . . . . . . . . . . . . . . . . . . . 25 103 11.1 Normative References . . . . . . . . . . . . . . . . . . . 25 104 11.2 Informative References . . . . . . . . . . . . . . . . . . 25 105 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 26 107 INTERNET DRAFT 109 1 Introduction 111 EVPN which is defined in RFC7432 which describes procedures for 112 BGP MPLS-based Ethernet VPNs(EVPN).This document defines the 113 methodologies for benchmarking performance of EVPN.EVPN has been 114 implemented with many varying designs in order to achieve their 115 intended network functionality.The scope of this document is to 116 provide methodologies for benchmarking evpn data,control plane mac 117 learning,mac flush,mac ageing,convergence,high availability,scale. 118 The methodologies defined for evpn can be used for benchmarking the 119 performance of PBB-EVPN.PBB-EVPN is defined in RFC 7623.It is 120 being deployed in provider network.The difference between PBB-EVPN 121 and EVPN is the former learns the customer mac in data plane the 122 later learns in control plane. 124 1.1 Terminology 126 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL 127 NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" 128 in this document are to be interpreted as described in RFC 2119 129 [RFC2119]. 131 2.Test Topology 133 Topology Diagram 135 Topology Diagram 137 | RT (Traffic Generator layer 2 traffic send to each sub interface it serve as sender and Reciever) 138 +----------+ 139 | | 140 | R1 | 141 | | 142 +----------+ 143 | 144 | 145 +----------+ 146 | | 147 | RR | 148 | | 149 +----------+-------------| 150 | | 151 | | 152 +----------+ +-----------+ 153 | | | MHPE2 | 154 | DUT | | | 155 | MHPE1 | | | 156 +----------+ +-----------+ 157 | | 158 +----------+------------ DUT and MHPE2 are running in Single Active mode 159 | | 160 | CE | 161 | | 162 +----------+------------ Traffic Generator 163 RT (Traffic Generator layer 2 traffic send to each sub interface it serve as sender and Reciever) 165 Topology Diagram 167 Figure 1 169 INTERNET DRAFT 171 2.1 Terminologies 173 MHPE Multi homed Provide Edge router. 175 RR Route Reflector 177 CE Customer Router/Devices 179 PE2 Provider Edge router 2 181 AA EVPN Terminologies AA All-Active 183 SA EVPN Terminologies SA Single-Active 185 RT Router Tester 187 3. Network 189 The network consists of 5 routers and 2 traffic generator 190 ports.DUT is acting as one of the MHPE to CE.The RR is acting as 191 route reflector and core router.R1 is a Single home router running 192 evpn. All four routers except CE are running mpls,bgp emulating a 193 provider scenario.CE is a dual home connected to DUT and MHPE2.The 194 testing will be done on DUT in order to bench mark the service. DUT 195 and the MHPE2 is running EVPN with SA/AA, CE will be configured 196 with layer 2 bridge.In AA EVPN there will be LAG running from CE to 197 both MHPE's.The DUT and other PE's will be running X EVI's (EVPN 198 instances) on X sub interfaces.The traffic generator will be 199 connected to R1 and the CE, which is capable of sending layer 2 200 frames.The traffic will be send either uni directional or bi 201 directional based on the benchmark parameters to be measured. 203 3.1 PBB-EVPN Network Setup 205 The network consists of 5 routers and 2 traffic generator 206 ports.DUT is acting as one of the MHPE to CE.The RR is acting as 207 route reflector and core router.R1 is a Single home router running 208 pbbevpn. All four routers except CE are running mpls,bgp emulating 209 a provider scenario.CE is a dual home connected to DUT and 210 MHPE2.DUT and MHPE2 is running PBB-EVPN with SA/AA with CE.CE will 211 be configured with layer 2 bridge.In AA PBB-EVPN there will be LAG 212 running from CE to both MHPE's.The DUT and other PE's will be 213 running X EVI's (PBB-EVPN instances) on X sub interfaces.The 214 traffic generator will be connected to R1 and the CE, which is 215 capable of sending layer 2 frames.The traffic will be send either 216 uni directional or bi directional based on the benchmark parameters 217 to be measured. 219 INTERNET DRAFT 221 4. Test Procedure 223 The test defined to bench mark the performance of EVPN mac learning 224 in control pland and data plane. Mac flush,High Availability, 225 convergence time in link failures,scale scenarios. 227 4.1 MAC Learning in Control plane and Data Plane 229 The MAC will be learned in data plane and control plane, test is to 230 measure the time taken to learn the "X" number of mac's in time "T" 231 sec. The data plane learning will from the locally connected 232 interface. The control plane learning is through BGP advertisements 233 from the remote PE. Let the local learning time be "T" and control 234 plane learning time be "T'". The data plane mac learning can be 235 measured using the parameters defined in RFC 2889 section 5.8. 237 4.2 MAC flush for locally learned and remote learned MAC 239 The time taken to flush the "X" locally learned mac, let it be "T1" 240 sec, once the traffic is stopped. The time taken to flush the remote 241 mac which is learned by control plane let it be "X" macs.The time 242 taken to flush the "X" remote macs, which is measured as "T2" sec. 244 4.3 High Availability 246 The traffic is flowing bi direction. The bgp is converged,consider 247 there are "X" numbers of macs learned locally and remotely. 248 Then traffic is flowing at "P" packets per sec. The traffic generator 249 is measuring the Tx and Rx packets, while the routing engine fail over there 250 should not any packet loss the router tester must show both "P" 251 packet per seconds. 253 4.4 Convergence Time 255 During any events like link failure, hard reset measure the time 256 taken to learn "X" mac's locally and remotely. 258 4.5 Scale 260 This is to measure the performance of DUT in scaling to "X" EVPN 261 instances. The measured parameters are CPU usage, memory 262 leak,crashes. 264 INTERNET DRAFT 266 4.6 SOAK 268 This test is used to measure the performance of DUT over a period of 269 time,with scaled configuration and traffic over a period of time 270 "T'". In each interval "t1" the parameters measured are CPU usage, 271 memory usage and crashes. 273 4.7 Measurement Statistics 275 The test is repeated for "N" times and the value is taken by 276 averaging the values. 278 5 Test Cases 280 The following tests are conducted to measure mac learning of local as 281 well as remote. 282 5.1.1 To check the time taken to learn the mac address in DUT 284 Objective: 286 To check the time taken to learn the mac address locally and time 287 taken to send the locally learned routes to peers. 289 a. Send X unicast frames from CE to MHPE1(DUT) working in SA 290 mode with different source and destination address, where DUT is 291 the DF so that it can forward the traffic.Measure the time 292 taken to learn these mac in forwarding table and in control plane. 293 The data plane learning is measured using RFC 2889 section 5.8. 294 Sending frames to the limit of bridge domain of particular EVI. 295 Measure the time taken to learn all X mac in data plane/hardware. 296 The Range of MAC is known from RT and this is verified in DUT. 298 b. Measure the time taken to send these X type 2 routes from 299 DUT to its peers. 301 Procedure: 303 Configure EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 304 running mpls,bgp,RR is acting as route reflector to R1,MHPE2 and 305 DUT.Once the bgp comes up check the DUT evpn table.For MH PE ESI must 307 INTERNET DRAFT 309 be configured per IFD/Interface.Using RT (traffic generator) to 310 send the traffic to the CE. The traffic is unidirectional.Since CE 311 is working in bridge mode,frames will be send to ingress sub 312 interface of DUT. 314 Measurement 316 The DUT mac table must learn the X macs in data plane in T time 317 frame.The DUT must send type 2 routes to remote router in T' time 318 frame.Repeat the test and plot the data.The data plane measurement 319 is taken by considering DUT as black box the range of X mac is 320 known from RT and the same is learned in DUT, the time to learn 321 that is measured. 323 PBB-EVPN To check the time taken to learn the mac address in DUT 325 Objective: 327 To check the time taken to learn the mac address locally. 329 a. Send X unicast frames from CE to MHPE1(DUT) working in SA 330 mode with different source and destination address, where DUT is 331 the DF so that it can forward the traffic. Measure the time 332 taken to learn these mac in forwarding table.The data plane 333 learning is measured using RFC 2889 section 5.8.Sending frames to 334 the limit of bridge domain of particular EVI.Measure the time 335 taken to learn all X mac in data plane/hardware.The Range of 336 MAC is known from RT and this is verified in DUT. 338 Procedure: 340 Configure PBB-EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 341 running mpls,bgp,RR is acting as route reflector to R1,MHPE2 and 342 DUT.Once the bgp comes up check the DUT evpn table.For MH PE ESI must 343 be configured per IFD/Interface.Using RT (traffic generator) to 344 send the traffic to the routers. 346 INTERNET DRAFT 348 Measurement 350 The DUT mac table must learn the X macs in data plane in T time 351 frame. Repeat the test and plot the data.The data plane measurement 352 is taken by considering DUT as black box the range of X mac is 353 known from RT and the same is learned in DUT, the time to learn 354 that is measured. 356 5.1.2. To check the time taken to learn X macs from remote peer by DUT 358 Objective: 360 Send X frames with different SA and DA to DUT from R1 using 361 traffic generator. Measure the time taken to learn these X 362 macs from remote peer and program the mac address table. 364 Procedure: 366 Configure EVPN EVI in R1,MHPE2,DUT. All 4 routers except CE are 367 running mpls,bgp,RR is acting as route reflector to R1,MHPE2 and 368 DUT.Once the bgp comes up check the DUT evpn table. For MH PE ESI 369 must be configured per IFD/Interface.Using RT(traffic generator) 370 send the traffic to R1.The traffic is uni directional.There wont 371 be any traffic flow from CE to DUT during this test. 373 Measurement: 375 The DUT mac table must learn the X mac address in T time frame. 376 Repeat these test and plot the data. 378 PBB-EVPN To check the time taken to learn X mac's from remote peer 379 by DUT. 381 Objective: 383 Send X frames with different SA and DA to DUT from R1 using 384 traffic generator.Measure the time taken to learn these X macs from 385 remote peer and program the mac address table. 387 Procedure: 389 Configure PBB-EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 390 running mpls,bgp,RR is acting as route reflector to R1,MHPE2 and 392 INTERNET DRAFT 394 DUT.Once the bgp comes up check the DUT pbb-evpn table. For MHPE ESI 395 must be configured per IFD/Interface.Using RT(traffic generator) 396 send the traffic to R1. The traffic is uni directional. There wont 397 be any traffic flow from CE to DUT during this test. 399 Measurement: 401 The DUT mac table must learn the X mac address in T time frame. 402 Repeat these test and plot the data. 404 The following test cases are executed to measure the time taken to 405 flush the mac table in case of an event 407 5.1.3. To check the time taken to flush the local entry due to CE link 408 failure and measure the relearning rate of X macs 410 Objective: 412 Send X frames with different SA and DA to DUT from CE using 413 traffic generator.Wait for a while to learn all X mac address. 414 Then fail the DUT CE link and measure the time taken to flush these 415 X macs from the mac table and from control plane. 417 Procedure: 419 Configure EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 420 running mpls,bgp,RR is acting as route reflector to R1,MHPE2 and DUT. 421 Once the bgp comes up check the DUT evpn table. For MH PE ESI must 422 be configured per IFD/Interface. Using RT(traffic generator) to send 423 the traffic to the routers.In this traffic will be only send from 424 CE. 426 Measurement: 428 Measure the time taken for flushing these X mac address. 429 Measure the time taken to relearn the X macs in other PE.Repeat 430 the test and plot the data. 432 PBB-EVPN To check the time taken to flush the local entry due to CE link failure 434 Objective: 436 Send X frames with different SA and DA to DUT from CE using 437 traffic generator.Wait for a while to learn all X mac address.Then 439 INTERNET DRAFT 441 fail the DUT CE link and measure the time taken to flush these X 442 macs from the mac table.Measure the time taken to relearn. 444 Procedure: 446 Configure PBB-EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 447 running mpls,bgp,RR is acting as route reflector to R1,MHPE2 and 448 DUT. Once the bgp comes up check the DUT pbb-evpn table. For MH PE 449 ESI must be configured per IFD/Interface. Using RT(traffic 450 generator) to send the traffic to the routers.In this traffic will 451 be only send from CE. 453 Measurement: 455 The DUT mac table must learn the X mac address and measure the 456 time taken for flushing these X mac address.Measure the time taken to relearn these 457 X macs. 459 5.1.4. To check the time taken by DUT to flush X routes learned from 460 remote PE after R1 traffic generator link failure 462 Objective: 464 Send X frames with different SA and DA to DUT from R1 using 465 traffic generator.Bring down the link between R1 and traffic 466 generator.Then measure the time taken to flush the DUT mac address 467 table. 469 Procedure: 471 Configure EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 472 running mpls,bgp,RR is acting as route reflector to R1,MHPE2 and 473 DUT.Once the bgp comes up check the DUT evpn table.For MH PE ESI must 474 be configured per IFD/Interface. Using RT(traffic generator) to 475 send the traffic to the R1.There wont be any traffic flowing to CE 476 from RT. 478 Measurement: 480 Measure the time taken to flush X remote routes from mac table 481 of DUT.Repeat the test and plot the data. 483 INTERNET DRAFT 485 PBB-EVPN To check the time taken by DUT to flush X macs learned 486 from remote PE after R1 traffic generator link failure 488 Objective: 490 Send X frames with different SA and DA to DUT from R1 using 491 traffic generator.Bring down the link between R1 and traffic 492 generator.Then measure the time taken to flush the DUT mac address 493 table.The remote macs will be learned by Data plane,but the B-MAC 494 will be learned by control plane. 496 Procedure: 498 Configure PBB-EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 499 running mpls,bgp,RR is acting as route reflector to R1,MHPE2 and 500 DUT. Once the bgp comes up check the DUT pbb-evpn table.For MH PE 501 ESI must be configured per IFD/Interface. Using RT(traffic 502 generator) to send the traffic to the routers.In this traffic will 503 be only send from CE. 505 Measurement: 507 Measure the time taken to flush X remote macs from mac table of 508 DUT. 510 The following test cases are executed to measure the mac ageing in 511 locally learned and remote mac learned by control plane. 513 5.1.5. To measure the mac ageing time. 515 Objective: 517 Send X frames with different SA and DA to DUT from CE using 518 traffic generator.Wait to learn all X mac address.Then stop the 519 traffic.Wait to see how long it takes to flush these mac entries 520 due to ageing. 522 Procedure: 524 Configure EVPN EVI in R1,MHPE2,DUT. All 4 routers except CE are 526 INTERNET DRAFT 528 running mpls,bgp,RR is acting as route reflector to R1,MHPE2 and 529 DUT.Once the bgp comes up check the DUT evpn table. For MH PE ESI 530 must be configured per IFD/Interface.Using RT(traffic generator) to 531 send the traffic to the routers.The traffic will be flowing from CE 532 to DUT.There wont be any traffic from R1. 534 Measurement: 536 Measure the time taken to flush X mac address due to ageing.Repeat 537 the test and plot the data. 539 PBB-EVPN To measure the mac ageing time. 541 Objective: 543 Send X frames with different SA and DA to DUT from CE using 544 traffic generator.Wait to learn all X mac address.Then stop the 545 traffic.Wait to see how long it takes to flush these mac entries 546 due to ageing. 548 Procedure: 550 Configure PBB-EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 551 running mpls,bgp,RR is acting as route reflector to R1,MHPE2 and 552 DUT.Once the bgp comes up check the DUT mac table. For MH PE ESI 553 must be configured per IFD/Interface. Using RT(traffic generator) to 554 send the traffic to the routers. 556 Measurement: 558 The DUT mac table must learn the X mac address, measure the time 559 taken for flushing these X mac address due to ageing. 561 INTERNET DRAFT 563 5.1.6. To check the time taken by DUT to age X routes learned from 564 remote PE after stopping the traffic at remote PE. 566 Objective: 568 Send X frames with different SA and DA to DUT from R1 using 569 traffic generator.After stopping the traffic at remote PE R1 570 traffic generator due to mac ageing it will withdraw its routes from 571 remote PE's.Measure the time taken to remove these macs from DUT 572 mac table. 574 Procedure: 576 Configure EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 577 running mpls,bgp,RR is acting as route reflector to R1,MHPE2 and 578 DUT.Once the bgp comes up check the DUT evpn table.For MH PE ESI must 579 be configured per IFD/Interface.Using RT(traffic generator) to send 580 the traffic to R1.There wont be any traffic from CE. 582 Measurement: 584 Measure the time taken to flush X remote macs from mac table of 585 DUT due to ageing.Repeat the test and plot the data. 587 PBB-EVPN To check the time taken by DUT to age X mac from remote 588 PE after stopping the traffic at remote PE. 590 Objective: 592 Send X frames with different SA and DA to DUT from R1 using 593 traffic generator.After stopping the traffic at remote PE(R1) traffic 594 generator.Measure the time taken to remove these macs from DUT mac 595 table. 597 Procedure: 599 Configure PBB-EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 600 running mpls,bgp,RR is acting as route reflector to R1,MHPE2 and 601 DUT.Once the bgp comes up check the DUT mac table.For MH PE ESI must 602 be configured per IFD/Interface.Using RT(traffic generator) to 603 send the traffic to the routers. 605 Measurement: 607 Measure the time taken to flush the remote macs from mac table of 608 DUT due to ageing. 610 INTERNET DRAFT 612 The following tests are executed to measure the convergence time in 613 case of an event or by lerning the mac without any external trigger. 615 5.1.7. To check the time taken by DUT to learn X routes from local and 616 X from remote and measure the time of flood from DUT. 618 Objective: 620 Send X frames with different SA and DA to DUT from R1 using 621 traffic generator.Send X frames with different SA and DA from 622 traffic generator connected to CE.The SA and DA of flows must be 623 complimentary to have unicast flows.Measure the time taken by the 624 DUT to learn 2X in mac table and in control plane.Measure the flood 625 time period of DUT. 627 Procedure: 629 Configure EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 630 running mpls,bgp,RR is acting as route reflector to R1,MHPE2 and 631 DUT.Once the bgp comes up check the DUT evpn table. For MH PE ESI 632 must be configured per IFD/Interface.Using RT(traffic generator) to 633 send the traffic to the routers.The traffic is bi directional. 635 Measurement: 637 Measure the time taken to learn 2X routes in control and mac 638 address table in DUT and measure the flood time of DUT.Repeat the 639 test and plot the data. 641 PBB-EVPN To check the time taken by DUT to learn X macs from 642 local and X from remote and measure the time of flood from 643 DUT. 645 Objective: 647 Send X frames with different SA and DA to DUT from R1 using 648 traffic generator. Send X frames with different SA and DA from 649 traffic generator connected to CE.The SA and DA of flows must be 650 complimentary to have unicast flows.Measure the time taken by the 651 DUT to learn 2X in mac table.Measure the flood time period of DUT. 653 INTERNET DRAFT 655 Procedure: 657 Configure PBB-EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 658 running mpls,bgp,RR is acting as route reflector to R1,MHPE2 and 659 DUT.Once the bgp comes up check the DUT evpn table.For MH PE ESI 660 must be configured per IFD/Interface.Using RT(traffic generator) to 661 send the traffic to the routers. 663 Measurement: 665 Measure the time taken to learn 2X mac address table in DUT. and 666 measure the flood time of DUT.Repeat the test and plot the data. 668 5.1.8. To measure the time taken to elect a new DF by adding a a MHPE. 670 Objective: 672 Send X frames from CE to DUT from traffic generator with 673 different SA and DA.Wait to learn all X mac address.Then add a new 674 router configured with evpn to the same ethernet segment.Configure 675 same ESI value which is configured in DUT and in MHPE1 on IFD.Then 676 the new DF election take place during that time there should not be 677 any loop and measure the time taken to come up the new DF.Measure 678 the time taken to elect the new DF. 680 Procedure: 682 Configure EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 683 running mpls,bgp,RR is acting as route reflector to R1,MHPE2 and 684 DUT.Once the bgp comes up check the DUT evpn table.For MHPE ESI 685 must be configured per IFD/Interface.Using RT(traffic generator) to 686 send the traffic to the routers. 688 Measurement: 690 Measure the time taken for new DF election in DUT and there should 692 INTERNET DRAFT 694 not be any loop or forwarding the BUM traffic back to the same 695 segment. 697 PBB-EVPN To measure the time taken to elect a new DF by adding a a 698 MHPE. 700 Objective: 702 Send X frames from CE to DUT from traffic generator with 703 different SA and DA.Wait to learn all X mac address.Then add a new 704 router to the same Ethernet segment by configuring the same ESI 705 value configured in DUT,MHPE2 in IFD.Then the new DF election take 706 place during that time there should not be any loop and measure the 707 time taken to come up the new DF.Measure the time taken to elect 708 the new DF. 710 Procedure: 712 Configure PBB-EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 713 running mpls,bgp,RR is acting as route reflector to R1,MHPE2 and 714 DUT.Once the bgp comes up check the DUT evpn table. For MH PE ESI 715 must be configured per IFD/Interface. Using RT(traffic generator) to 716 send the traffic to the routers. 718 Measurement: 720 Measure the time taken for new DF election in DUT and there should 721 not be any loop or forwarding the BUM traffic back to the same 722 segment. 724 INTERNET DRAFT 726 5.2 High Availability 728 These tests are conducted to check after an event there wont be 729 any change in functionality. 731 5.2.1 To check the whether there is traffic loss due to routing engine 732 failover for redundancy test. 734 Objective: 736 Send X frames from CE to DUT from traffic generator with 737 different SA and DA. Send X frames from traffic generator to R1 738 with different SA and DA so that 2X mac address will be learned in 739 DUT. There is a bi directional traffic flow with X pps in each 740 direction. Then do a routing engine failover. 742 Procedure: 744 Configure EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 745 running mpls,bgp,RR is acting as route reflector to R1,MHPE2 and 746 DUT.Once the bgp comes up check the DUT evpn table.For MH PE ESI must 747 be configured per IFD/Interface.Using RT(traffic generator) to send 748 the traffic to the routers. 750 Measurement: 752 There should be 0 traffic loss which is the ideal case,No 753 change in the DF role.No withdraw of any routes. 755 PBB-EVPN To check the whether there is traffic loss due to routing 756 engine failover for redundancy test. 758 Objective: 760 Send X frames from CE to DUT from traffic generator with 761 different SA and DA.Send X frames from traffic generator to R1 with 762 different SA and DA so that 2X mac address will be learned in DUT. 763 There is a bi directional traffic flow with X pps in each 764 direction.Then do a routing engine failover. 766 Procedure: 768 Configure PBB-EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 770 INTERNET DRAFT 772 running mpls,bgp,RR is acting as route reflector to R1,MHPE2 and 773 DUT.Once the bgp comes up check the DUT evpn table.For MH PE ESI must 774 be configured per IFD/Interface. Using RT(traffic generator) to 775 send the traffic to the routers. 777 Measurement: 779 There should be 0 packet loss loss which is a ideal case,No change 780 in the DF role. No withdraw of any routes. 782 5.3 ARP/ND Scaling 784 These tests are conducted to check the scaling parameter of arp/ND 785 of the DUT. 787 EVPN: To check the ARP/ND scale of the DUT when the gateway is 788 configured. 790 Objective: 792 Send X arp/icmpv6 request from RT to DUT,that with different 793 sender ip/ipv6 address to the same target gateway ip address. 795 Procedure: 797 Configure EPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 798 running mpls,bgp,RR is acting as route reflector to R1,MHPE2 and 799 DUT.Once the bgp comes up check the DUT evpn table.For MH PE ESI must 800 be configured per IFD/Interface. Using RT(traffic generator) to 801 send the arp request to the DUT which has gateway configured. 803 Measurement: 805 The DUT must learn X mac+ip/mac+ipv6 and it must advertise the X 806 mac+ip/ mac+ipv6 to the remote router. 808 INTERNET DRAFT 810 6. Scale 811 6.1. To Scale the DUT to N EVI and clear bgp in DUT with out traffic. 813 Objective: 815 INTERNET DRAFT 817 The main purpose of the test the DUT performance to scale N EVI's. 818 Then clear bgp neighbor. There should not be any loss of routes or 819 any crashes. 821 Procedure: 823 Configure EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 824 running mpls,bgp,RR is acting as route reflector to R1,MHPE2 and 825 DUT.Once the bgp comes up check the DUT evpn table.For MHPE,DUT ESI 826 must be configured per IFD/Interface. Using RT(traffic generator) 827 to send the traffic to the routers. 829 Measurement 831 There should not be any loss of route types 1,2,3 and 4 in DUT. 833 PBB-EVPN To Scale the DUT to N PBB-EVPN instances and clear bgp in 834 DUT without traffic. 836 Objective: 838 The main purpose of the test the DUT performance to scale N PBB- 839 EVPN instances.Then clear bgp neighbor.There should not be any loss 840 of routes or any crashes. 842 Procedure: 844 Configure PBB-EVPN instances in R1,MHPE2,DUT.All 4 routers except 845 CE are running mpls,bgp,RR is acting as route reflector to R1,MHPE2 846 and DUT.Once the bgp comes up check the DUT evpn table.For MH PE 847 ESI must be configured per IFD/Interface.Using RT(traffic 848 generator) to send the traffic to the routers. 850 Measurement 852 There should not be any loss of route types 2,3 and 4 in DUT. 854 INTERNET DRAFT 856 6.2. To Scale the DUT to N EVI and clear bgp in DUT with 857 traffic.Measure the convergence time 859 Objective: 861 The main purpose of the test the DUT performance to scale N EVI's 862 with traffic. Then clear bgp neighbor.There should not be any loss 863 of routes or any crashes. Send F frames from CE to DUT from 864 traffic generator with different SA and DA for N EVI's. Send F 865 frames from traffic generator to R1 with different SA and DA for N 866 EVI's.so that 2F number of mac address will be learned in DUT. 867 There is a bi directional traffic flow with F pps in each 868 direction. Then clear the bgp nei in DUT after the bgp comes up and 869 started learning the routes, measure the time taken to learn all 870 2F mac routes. 872 Procedure: 874 Configure EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 875 running mpls,bgp,RR is acting as route reflector to R1,MHPE2 and 876 DUT.Once the bgp comes up check the DUT evpn table.For MH PE ESI must 877 be configured per IFD/Interface. Using RT(traffic generator) to 878 send the traffic to the routers. 880 Measurement: 882 The DUT must learn all 2F mac address.Measure the time taken to 883 learn 2F mac in DUT, measure the flood traffic time "t" of DUT 885 PBB-EVPN To Scale the DUT to "N" PBB-EVPN instances and clear bgp 886 in DUT with traffic.Measure the convergence time 888 Objective: 890 The main purpose of the test the DUT performance to scale N pbb- 891 evpn instances with traffic.Then clear bgp neighbor.There should 892 not be any loss of routes or any crashes.Send F frames from CE to 893 DUT from traffic generator with different SA and DA for N pbb-evpn 894 instances.Send F frames from traffic generator to R1 with different 895 SA and DA for N pbb-evpn instances.so that F mac address will be 896 learned in DUT.There is a bi directional traffic flow with F pps in 897 each direction.Then clear the bgp nei in DUT after the bgp comes up 898 and started learning the routes, measure the time taken to learn 899 all 2F macs in DUT. 901 INTERNET DRAFT 903 Procedure: 905 Configure PBB-EVPN instances in R1,MHPE2,DUT.All 4 routers except 906 CE are running mpls,bgp,RR is acting as route reflector to R1,MHPE2 907 and DUT.Once the bgp comes up check the DUT evpn table.For MH PE 908 ESI must be configured per IFD/Interface. Using RT(traffic 909 generator) to send the traffic to the routers. 911 Measurement: 913 The DUT must learn all 2F mac address.Measure the time taken 914 to learn 2F mac in DUT, measure the flood traffic time "t" of DUT 916 7. Soak Test 918 7.1. To Scale the DUT to N EVI in DUT with traffic and run the set up 919 for 24hrs 921 Objective: 923 The main purpose of the test the DUT performance to scale N EVI's 924 with traffic. Then clear bgp neighbor. There should not be any 925 loss of routes or any crashes. Send F frames from CE to DUT from 926 traffic generator with different SA and DA for N EVI's. Send F 927 frames from traffic generator to R1 with different SA and DA for N 928 EVI's.so that 2F mac address will be learned in DUT. There is a bi 929 directional traffic flow with F pps in each direction.Keep the 930 setup up and running for 24 hrs,take hourly CPU utilization,memory 931 usage. 933 Procedure: 935 Configure EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 936 running mpls,bgp,RR is acting as route reflector to R1,MHPE2 and 937 DUT.Once the bgp comes up check the DUT evpn table.For MH PE ESI must 938 be configured per IFD/Interface. Using RT(traffic generator) to 939 send the traffic to the routers. 941 Measurement: 943 Take the hourly reading of CPU,process memory.There should not be 944 any leak,crashes,CPU spikes. 946 INTERNET DRAFT 948 PBB-EVPN To Scale the DUT to N PBB-EVPN instances in DUT with 949 traffic and run the set up for 24hrs 951 Objective: 953 The main purpose of the test the DUT performance to scale N EVI's 954 with traffic. Then clear bgp neighbor. There should not be any 955 loss of routes or any crashes. Send F frames from CE to DUT from 956 traffic generator with different SA and DA for N EVI's. Send F 957 frames from traffic generator to R1 with different SA and DA for N 958 EVI's.so that 2F mac address will be learned in DUT. There is a bi 959 directional traffic flow with F pps in each direction.Keep the 960 setup up and running for 24 hrs,take hourly CPU utilization,memory 961 usage. 963 Procedure: 965 Configure PBB-EVPN instances in R1,MHPE2,DUT.All 4 routers except 966 CE are running mpls,bgp,RR is acting as route reflector to R1,MHPE2 967 and DUT.Once the bgp comes up check the DUT evpn table.For MH PE 968 ESI must be configured per IFD/Interface. Using RT(traffic 969 generator) to send the traffic to the routers. 971 Measurement: 973 Take the hourly reading of CPU,process memory.There should not be 974 any leak,crashes,CPU spikes. 976 INTERNET DRAFT 978 8. Acknowledgements 980 We would like to thank Fioccola Giuseppe of Telecom Italia 981 reviewing our draft and commenting it. 982 9. IANA Considerations 984 This memo includes no request to IANA. 985 10. Security Considerations 987 There is no additional consideration from RFC 6192. 988 11 References 990 11.1 Normative References 992 [RFC2119] Bradner, S., "Key words for use in RFCs to 993 Indicate Requirement Levels", BCP 14, RFC 2119, 994 DOI 10.17487/RFC2119, June 1997,. 997 [RFC2544] Bradner, S. and J. McQuaid, "Benchmarking 998 Methodology for Network Interconnect Devices", RFC 999 2544,DOI 10.17487/RFC2544, June 1999,. 1002 [RFC2889] R.Mandeville and J. Perser "Benchmarking Methodology 1003 for LAN Switching Devices" 1005 11.2 Informative References 1007 [RFC7432] Sajassi, A., Ed., Aggarwal, R., Bitar, N., 1008 Isaac, A.,Uttaro, J., Drake, J., and W.Henderickx, 1009 "BGP MPLS-Based Ethernet VPN", 1010 RFC 7432, DOI 10.17487/RFC7432, February 1011 2015, . 1013 [RFC7623] Sajassi, A., Ed., Aggarwal, R., Bitar, N., 1014 Isaac, A.,J., Drake, J., and W. Henderickx, 1015 " Provider Backbone Bridging Combined with Ethernet 1016 VPN(PBB-EVPN)",RFC 7623,10.17487/RFC7623,September 2015 1017 . 1019 INTERNET DRAFT 1021 Authors' Addresses 1023 Appendix A. Appendix 1025 Authors' Addresses 1027 Sudhin Jacob (editor) 1028 Juniper Networks 1029 Bangalore 1030 India 1032 Phone: +91 8061212543 1033 Email: sjacob@juniper.net 1034 sudhinjacob@rediffmail.com 1036 Kishore Tiruveedhula 1037 Juniper Networks 1038 10 Technology Park Dr 1039 Westford, MA 01886 1040 USA 1042 Phone: +1 9785898861 1043 Email: kishoret@juniper.net