idnits 2.17.1 draft-kishjac-bmwg-evpntest-10.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 : ---------------------------------------------------------------------------- ** There are 2 instances of too long lines in the document, the longest one being 62 characters in excess of 72. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == Line 164 has weird spacing: '... Tester sendi...' == The document doesn't use any RFC 2119 keywords, yet seems to have RFC 2119 boilerplate text. -- The document date (June 24, 2018) is 2126 days in the past. Is this intentional? Checking references for intended status: Proposed 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 943, but no explicit reference was found in the text == Unused Reference: 'RFC2899' is defined on line 948, but no explicit reference was found in the text == Unused Reference: 'RFC7432' is defined on line 954, but no explicit reference was found in the text == Unused Reference: 'RFC7623' is defined on line 959, 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 2899 Summary: 3 errors (**), 0 flaws (~~), 7 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Internet Engineering Task Force S. Jacob, Ed. 3 Internet-Draft K. Tiruveedhula 4 Intended status: Standards Track Juniper Networks 5 Expires: December 26, 2018 June 24, 2018 7 Benchmarking Methodology for EVPN and PBB-EVPN 8 draft-kishjac-bmwg-evpntest-10 10 Abstract 12 This document defines methodologies for benchmarking EVPN and PBB- 13 EVPN performance. EVPN is defined in RFC 7432, and is being deployed 14 in Service Provider networks. Specifically this document defines the 15 methodologies for benchmarking EVPN/PBB-EVPN convergence, data plane 16 performance, and control plane performance. 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 https://datatracker.ietf.org/drafts/current/. 28 Internet-Drafts are draft documents valid for a maximum of six months 29 and may be updated, replaced, or obsoleted by other documents at any 30 time. It is inappropriate to use Internet-Drafts as reference 31 material or to cite them other than as "work in progress." 33 This Internet-Draft will expire on December 26, 2018. 35 Copyright Notice 37 Copyright (c) 2018 IETF Trust and the persons identified as the 38 document authors. All rights reserved. 40 This document is subject to BCP 78 and the IETF Trust's Legal 41 Provisions Relating to IETF Documents 42 (https://trustee.ietf.org/license-info) in effect on the date of 43 publication of this document. Please review these documents 44 carefully, as they describe your rights and restrictions with respect 45 to this document. Code Components extracted from this document must 46 include Simplified BSD License text as described in Section 4.e of 47 the Trust Legal Provisions and are provided without warranty as 48 described in the Simplified BSD License. 50 Table of Contents 52 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 53 1.1. Requirements Language . . . . . . . . . . . . . . . . . . 3 54 1.2. Terminologies . . . . . . . . . . . . . . . . . . . . . . 3 55 2. Test Topology . . . . . . . . . . . . . . . . . . . . . . . . 4 56 3. Test Cases . . . . . . . . . . . . . . . . . . . . . . . . . 6 57 3.1. How long it takes to learn local mac address in EVPN . . 6 58 3.2. How long it takes to learn local mac address in PBB EVPN 6 59 3.3. How long it takes to learn the remote macs . . . . . . . 7 60 3.4. PBB-EVPN How long it takes to learn the mac from remote 61 peer . . . . . . . . . . . . . . . . . . . . . . . . . . 8 62 3.5. How long it takes to flush the local macs due to CE link 63 flap and measure the relearning rate of MACs . . . . . . 8 64 3.6. PBB-EVPN how long it takes to flush the local macs and 65 measure the relearning rate of macs during PE-CE link 66 flap . . . . . . . . . . . . . . . . . . . . . . . . . . 9 67 3.7. How long it takes to flush the remote macs, due to remote 68 link failure. . . . . . . . . . . . . . . . . . . . . . . 10 69 3.8. PBB-EVPN How long it takes to flush the remote macs due 70 to remote link failure . . . . . . . . . . . . . . . . . 10 71 3.9. To measure the MAC aging time. . . . . . . . . . . . . . 11 72 3.10. PBB-EVPN To measure the MAC aging time. . . . . . . . . . 12 73 3.11. How long it takes to age out the remote macs . . . . . . 12 74 3.12. PBB-EVPN How long it takes to age out the remote macs. . 13 75 3.13. How long it takes to learn both local and remote macs. . 14 76 3.14. PBB-EVPN How long it takes to learn both local and remote 77 macs . . . . . . . . . . . . . . . . . . . . . . . . . . 14 78 4. High Availability . . . . . . . . . . . . . . . . . . . . . . 15 79 4.1. To Record the whether there is traffic loss due to 80 routing engine failover for redundancy test. . . . . . . 15 81 4.2. PBB-EVPN To Record the whether there is traffic loss due 82 to routing engine failover for redundancy test . . . . . 16 83 5. ARP/ND Scale . . . . . . . . . . . . . . . . . . . . . . . . 16 84 5.1. To find ARP/ND scale . . . . . . . . . . . . . . . . . . 16 85 6. Scale . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 86 6.1. To Measure the scale limit of DUT with trigger (Scale 87 without traffic) . . . . . . . . . . . . . . . . . . . . 17 88 6.2. PBB-EVPN To measure the scale limit with trigger. . . . . 17 89 6.3. To measure the convergence time of DUT with scale and 90 traffic. . . . . . . . . . . . . . . . . . . . . . . . . 18 91 6.4. .PBB-EVPN To measure the convergence time of DUT with 92 scale and traffic. . . . . . . . . . . . . . . . . . . . 19 93 7. SOAK Test . . . . . . . . . . . . . . . . . . . . . . . . . . 19 94 7.1. To Measure the stability of the DUT with scale and 95 traffic. . . . . . . . . . . . . . . . . . . . . . . . . 19 96 7.2. PBB-EVPN to measure the stability of DUT with scale and 97 traffic. . . . . . . . . . . . . . . . . . . . . . . . . 20 99 8. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 21 100 9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 21 101 10. Security Considerations . . . . . . . . . . . . . . . . . . . 21 102 11. References . . . . . . . . . . . . . . . . . . . . . . . . . 21 103 11.1. Normative References . . . . . . . . . . . . . . . . . . 21 104 11.2. Informative References . . . . . . . . . . . . . . . . . 21 105 Appendix A. Appendix . . . . . . . . . . . . . . . . . . . . . . 22 106 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 22 108 1. Introduction 110 EVPN is defined in RFC 7432, and describes BGP MPLS- based Ethernet 111 VPNs (EVPN). PBB-EVPN is defined in RFC 7623, discusses how Ethernet 112 Provider backbone Bridging can be combined with EVPNs to provide a 113 new/combined solution. This draft defines methodologies that can be 114 used to benchmark both RFC 7432 and RFC 7623 solutions. Further, 115 this draft provides methodologies for benchmarking the performance of 116 EVPN data and control planes, MAC learning, MAC flushing, MAC ageing, 117 convergence, high availability, and scale. 119 1.1. Requirements Language 121 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 122 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 123 document are to be interpreted as described in RFC 2119 [RFC2119]. 125 1.2. Terminologies 127 MHPE Multi homed Provide Edge router. 129 RR Route Reflector. 131 P Provider Router. 133 CE Customer Router/Devices/Switch. 135 MHPE2 Multi homed Provider Edge router 2. 137 MHPE1 Multi homed Provider Edge router 1. 139 SHPE3 Single homed Provider Edge Router 3. 141 AA EVPN Terminologies AA All-Active. 143 SA EVPN Terminologies SA Single-Active. 145 RT Router Tester. 147 Sub Interface Each physical Interfaces is subdivided in to Logical 148 units. 150 EVI EVPN Instances which will be running on sub interface or physical 151 port of the provider Edge routers. 153 DF Designated Forwarder. 155 ESI Ethernet Segment Identifier. 157 2. Test Topology 159 EVPN/PBB-EVPN Services running on R1, MHPE1 and MHPE2 in Single 160 Active Mode: 162 Topology Diagram 164 | [Traffic Generator ] Router Tester sending layer 2 bi directional traffic sender/receiver 165 +----------+ 166 | | 167 | R1 | 168 | | 169 +----------+ 170 | 171 |Core link 172 +----------+ 173 | | 174 | RR | 175 | | Route Reflector/Core router 176 +----------+-------------| 177 | | 178 | Core links | 179 +----------+ +-----------+ 180 | | | MHPE2 | 181 | DUT | | | 182 | MHPE1 | | | 183 +----------+ +-----------+ 184 | PE-CE link | 185 +----------+------------ 186 | | 187 | CE | 188 | layer2 | 189 |bridge | 190 +----------+------------ [Traffic Generator](Router Tester sending bi directional layer 2 traffic with different VLAN sender/receiver) 192 Topology Diagram 194 Figure 1 196 There are five routers in the topology. R1, RR/P, MHPE1 and MHPE2 197 emulating a service provider network. CE is a customer device 198 connected to MHPE1 and MHPE2, it is configured with bridge domains in 199 different vlans. The Router tester is connected to R1 and CE which 200 send layer 2 traffic for configured vlans. MHPE1,MHPE2,RR/P,R1 run 201 MPLS. The EVPN/PBB-EVPN services are running on MHPE1,MHPE2 and R1. 202 The MHPE1 acts DUT. The RT will act as sender and receiver. The 203 measurement will be taken in DUT. 205 3. Test Cases 207 The following tests are conducted to measure the time taken to learn 208 the "X" number of MAC's locally in EVI . The data plane learning of 209 MAC will happen locally from connected interface. The control plane 210 learning of MAC is through BGP advertisements from the remote 211 PE(SHPE3). The control plane learning of "X" MAC. The data plane 212 MAC learning can be measured using the parameters defined in RFC 2889 213 section 5.8. 215 3.1. How long it takes to learn local mac address in EVPN 217 Objective: 219 To Record the time taken to learn the MAC address locally in DUT. 221 Procedure: 223 Configure EVPN EVI in R1, MHPE2, DUT. All 4 routers except CE are 224 running MPLS, BGP, RR is acting as route reflector to R1,MHPE2 and 225 DUT. For MH PE ESI must be configured per IFD/Interface. Using RT 226 (traffic generator)to send the traffic to the CE. The traffic is 227 unidirectional. Since CE is working in bridge mode, frames will be 228 send to ingress sub interface of DUT. The BGP must be established in 229 R1, MHPE1(DUT), RR, MHPE2. Send "X" unicast frames from CE to 230 MHPE1(DUT) working in SA mode with "X" different source and 231 destination address from RT. The DUT must learn these "X" macs in 232 data plane. 234 Measurement : 236 Measure the time taken to learn "X" MACs in DUT evpn mac table. The 237 data plane measurement is taken by considering DUT as black box the 238 range of X MAC is known from RT and the same must be learned in DUT, 239 the time taken to learn "X" macs is measured. 241 Repeat these test and plot the data. The test is repeated for "N" 242 times and the values are collected. The mac learning time is 243 calculated by averaging the values obtained from "N" samples. 245 Mac learning in sec = (T1+T2+..Tn/N) 247 3.2. How long it takes to learn local mac address in PBB EVPN 249 Objective: 251 To Record the time taken to learn the MAC address locally. 253 Procedure: 255 Configure PBB-EVPN EVI in R1,MHPE2,DUT. All 4 routers except CE are 256 running MPLS, BGP, RR is acting as route reflector to R1,MHPE2 and 257 DUT. Once the BGP comes up. Record the DUT PBB-EVPN table. For MH 258 PE ESI must be configured per IFD/Interface. From RT (traffic 259 generator) send the traffic to the DUT. The BGP must be established 260 in R1,MHPE1 (DUT), RR, MHPE2. The traffic is unidirectional. Since 261 CE is working in bridge mode, frames will be send to ingress sub 262 interface of DUT. 264 Send "X" unicast frames from CE to MHPE1(DUT) working in SA mode with 265 "X" different source and destination address from RT. The DUT must 266 learn "X" macs in data plane. 268 Measurement : 270 Measure the time taken by the DUT to learn the "X" MACs in the data 271 plane. The data plane measurement is taken by considering DUT as 272 black box the range of "X" MAC is known from RT and the same must be 273 learned in DUT, the time taken to learn "X" MAC is measured. Repeat 274 these test and plot the data. The test is repeated for "N" times and 275 the values are collected. The mac learning time is calculated by 276 averaging the values obtained from "N" samples. 278 Mac learning in sec = (T1+T2+..Tn/N) 280 3.3. How long it takes to learn the remote macs 282 Objective: 284 To Record the time taken to learn the remote macs. 286 Procedure: 288 Configure EVPN EVI in R1,MHPE2,DUT. All 4 routers except CE are 289 running MPLS,BGP,RR is acting as route reflector to R1,MHPE2 and DUT. 290 Record the DUT EVPN table. For MH PE ESI must be configured per IFD/ 291 Interface. Using RT(traffic generator) send the traffic to R1.The 292 traffic is uni directional. There wont be any traffic flow from CE 293 to DUT during this test. The BGP must be in established state. The 294 MACS learned in R1 will be advertised to DUT by BGP. 296 Send X frames with X different SA and DA to R1 from RT. R1 will 297 advertise these locally learned macs to MHPE1 and MHPE2 via control 298 plane.Measure the time taken to learn these X MACs from remote peer 299 in DUT EVPN MAC address table.The DUT and MHPE2 are running SA mode. 301 Measurement : 303 Measure the time taken by the DUT to learn the "X" MACs in the data 304 plane.Repeat these test and plot the data.The test is repeated for 305 "N" times and the values are collected.The mac learning time is 306 calculated by averaging the values obtained from "N" samples. 308 Mac learning in sec = (T1+T2+..Tn/N) 310 3.4. PBB-EVPN How long it takes to learn the mac from remote peer 312 Objective: 314 To Record the time taken to learn the remote macs. 316 Procedure: 318 Configure PBB-EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 319 running MPLS,BGP,RR is acting as route reflector to R1,MHPE2 and DUT. 320 Record the DUT PBB-EVPN table. For MHPE ESI must be configured per 321 IFD/Interface. Using RT(traffic generator) send the traffic to R1. 322 The traffic is uni directional.There wont be any traffic flow from CE 323 to DUT during this test.The BGP must be in established state. 325 Send X frames with X different SA and DA to R1 from RT.These macs 326 will be flooded to MHPE1 and MHPE2 by R1.The DUT and MHPE2 are 327 running SA mode. 329 Measurement : 331 Measure the time taken to learn X mac address in DUT mac table. 332 Repeat these test and plot the data.The test is repeated for "N" 333 times and the values are collected.The mac learning time is 334 calculated by averaging the values obtained by "N" samples. 336 Mac learning in sec = (T1+T2+..Tn/N) 338 3.5. How long it takes to flush the local macs due to CE link flap and 339 measure the relearning rate of MACs 341 Objective: 343 To record the time taken to flush the mac learned locally and the 344 time taken to relearn the same amount of macs. 346 Procedure: 348 Configure EVPN EVI in R1,MHPE2,DUT. All 4 routers except CE are 349 running MPLS,BGP,RR is acting as route reflector to R1,MHPE2 and DUT. 350 Once the BGP is established. Record the DUT EVPN table. For MH PE 351 ESI must be configured per IFD/Interface. Using RT(traffic 352 generator) send the traffic. In this scenario traffic will be only 353 send from CE side. 355 Send X frames with X different SA and DA to DUT from CE using traffic 356 generator. Wait till the MHPE1 learns all X MAC address. Then fail 357 the MHPE1 CE link and measure the time taken to flush these X MACs 358 from the EVPN MAC table. Bring up the link which was made Down(the 359 link between MHPE1 and CE).Measure time taken to relearn it. The DUT 360 and MHPE2 are running SA mode. 362 Measurement : 364 Measure the time taken for flushing these X MAC address. Measure the 365 time taken to relearn the X MACs in DUT. Repeat these test and plot 366 the data. The test is repeated for "N" times and the values are 367 collected. The flush and the relearning time is calculated by 368 averaging the values obtained by "N" samples. 370 Flush time for X Macs in sec = (T1+T2+..Tn/N) Relearning time for X 371 macs in sec = (T1+T2+..Tn/N) 373 3.6. PBB-EVPN how long it takes to flush the local macs and measure the 374 relearning rate of macs during PE-CE link flap 376 Objective: 378 To record the time taken to flush the mac learned locally and the 379 time taken to relearn the same amount of macs. 381 Procedure: 383 Configure PBB-EVPN EVI in R1,MHPE2,DUT. All 4 routers except CE are 384 running MPLS,BGP,RR is acting as route reflector to R1,MHPE2 and DUT. 385 Once the BGP is established. Record the DUT PBB-EVPN table. For MH 386 PE ESI must be configured per IFD/Interface. Using RT(traffic 387 generator) send the traffic to the CE. The traffic is uni 388 directional. 390 Send X frames with X different SA and DA to DUT from CE using traffic 391 generator. Wait till the MHPE1 learn all X MAC address. Then fail 392 the MHPE1 CE link and measure the time taken to flush these X MACs 393 from the PBB-EVPN MAC table. Then bring up the link. Measure the 394 time taken to relearn X MACS. The DUT and MHPE2 are running SA mode. 396 Measurement : 398 Measure the time taken for flushing these X MAC address. Measure the 399 time taken to relearn the X MACs in DUT. Repeat these test and plot 400 the data. The test is repeated for "N" times and the values are 401 collected. The flush and the relearning time is calculated by 402 averaging the values obtained by "N" samples. 404 Flush time for X Macs in sec = (T1+T2+..Tn/N) Relearning time for X 405 macs in sec = (T1+T2+..Tn/N) 407 3.7. How long it takes to flush the remote macs, due to remote link 408 failure. 410 Objective: 412 To record the time taken to flush the remote mac learned in DUT 413 during remote link failure. 415 Procedure: 417 Configure EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 418 running MPLS,BGP,RR is acting as route reflector to R1,MHPE2 and DUT. 419 Once the BGP is established Record the DUT EVPN table. For MHPE ESI 420 must be configured per IFD/Interface. Using RT(traffic generator) 421 send the traffic to R1.There wont be any traffic flowing to CE from 422 RT. 424 Send X frames with X different SA and DA to DUT from R1 using traffic 425 generator. Bring down the link between R1 and traffic generator. 426 Then measure the time taken to flush the DUT EVPN MAC table. The DUT 427 and MHPE2 are running SA mode. 429 Measurement : 431 Measure the time taken to flush X remote MACs from EVPN MAC table of 432 DUT. Repeat these test and plot the data. The test is repeated for 433 "N" times and the values are collected. The flush rate is calculated 434 averaging the values obtained by "N" samples. 436 Flush time for X Macs in sec = (T1+T2+..Tn/N) 438 3.8. PBB-EVPN How long it takes to flush the remote macs due to remote 439 link failure 441 Objective: 443 To record the time taken to flush the remote mac learned in DUT 444 during remote link failure. 446 Procedure: 448 Configure PBB-EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 449 running MPLS,BGP,RR is acting as route reflector to R1,MHPE2 and DUT. 450 Once the BGP is established Record the DUT PBB-EVPN MAC table. For 451 MHPE ESI must be configured per IFD/Interface. Using RT(traffic 452 generator) send the traffic to R1.In this scenario traffic will be 453 flowing only from R1. 455 Send X frames with X different SA and DA to DUT from R1 using traffic 456 generator. Bring down the link between R1 and traffic generator. 457 Then measure the time taken to flush the DUT PBB-EVPN MAC address 458 table. The remote MACs will be learned by Data plane, but the B-MAC 459 will be learned by control plane. The DUT and MHPE2 are running SA 460 mode. 462 Measurement : 464 Measure the time taken to flush X remote MACs from PBB-EVPN MAC table 465 of DUT. Repeat these test and plot the data. The test is repeated 466 for "N" times and the values are collected. The flush rate is 467 calculated by averaging the values obtained by "N" samples. 469 Flush time for X Macs in sec = (T1+T2+..Tn/N) 471 3.9. To measure the MAC aging time. 473 Objective: 475 To measure the mac aging time. 477 Procedure: 479 Configure EVPN EVI in R1,MHPE2,DUT. All 4 routers except CE are 480 running MPLS,BGP,RR is acting as route reflector to R1,MHPE2 and DUT 481 .Once the BGP is established. Record the DUT EVPN table. For MHPE 482 ESI must be configured per IFD/Interface. Using RT(traffic 483 generator), send the traffic to the DUT. The traffic will be flowing 484 from CE to DUT. There wont be any traffic from R1. 486 Send X frames with X different SA and DA to DUT from CE using traffic 487 generator. Wait till X MAC address are learned. Then stop the 488 traffic. Record the time taken to flush X MACS from DUT EVPN MAC 489 table due to aging. The DUT and MHPE2 are running SA mode 490 Measurement : 492 Measure the time taken to flush X MAC address due to aging. Repeat 493 these test and plot the data. The test is repeated for "N" times and 494 the values are collected. The aging is calculated averaging the 495 values obtained by "N" samples. 497 Aging time for X Macs in sec = (T1+T2+..Tn/N) 499 3.10. PBB-EVPN To measure the MAC aging time. 501 Objective: 503 To measure the mac aging time. 505 Procedure: 507 Configure PBB-EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 508 running MPLS,BGP,RR is acting as route reflector to R1,MHPE2 and 509 DUT.Once the BGP is established. Record the DUT PBB-EVPN MAC table. 510 For MH PE ESI must be configured per IFD/Interface. Using RT(traffic 511 generator) send the traffic to the DUT. The traffic is 512 unidirectional flowing from CE to DUT. 514 Send X frames with X different SA and DA to DUT from CE using traffic 515 generator. Wait till X MAC address are learned in DUT PBB- EVPN MAC 516 table. Then stop the traffic. Record the time taken to flush X MAC 517 entries due to aging. The DUT and MHPE2 running in SA mode 519 Measurement : 521 Measure the time taken to flush X MAC address due to aging. Repeat 522 these test and plot the data. The test is repeated for "N" times and 523 the values are collected. The aging is calculated by averaging the 524 values obtained by "N" samples. 526 Aging time for X Macs in sec = (T1+T2+..Tn/N) 528 3.11. How long it takes to age out the remote macs 530 Objective: 532 To measure the remote mac aging time. 534 Procedure: 536 Configure EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 537 running MPLS,BGP,RR is acting as route reflector to R1,MHPE2 and DUT. 539 Once the BGP is established. Record the DUT EVPN table. For MH PE 540 ESI must be configured per IFD/Interface. Using RT(traffic 541 generator) send the traffic to R1.There wont be any traffic from CE. 543 Send X frames with X different SA and DA to DUT from R1 using traffic 544 generator. Stop the traffic at remote PE R1.Due to MAC aging R1 will 545 withdraw its routes from DUT and MHPE2. Measure the time taken to 546 remove these MACs from DUT EVPN MAC table. DUT and MHPE2 are running 547 in SA mode 549 Measurement : 551 Measure the time taken to flush X remote MACs learned in DUT EVPN MAC 552 table due to aging. Repeat these test and plot the data. The test 553 is repeated for "N" times and the values are collected. The aging is 554 calculated by averaging the values obtained by "N" samples. 556 Aging time for X Macs in sec = (T1+T2+..Tn/N) 558 3.12. PBB-EVPN How long it takes to age out the remote macs. 560 Objective: 562 To measure the remote mac aging time. 564 Procedure: 566 Configure PBB-EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 567 running MPLS,BGP,RR is acting as route reflector to R1,MHPE2 and DUT. 568 Once the BGP is established. Record the DUT MAC table. For MH PE 569 ESI must be configured per IFD/Interface. Using RT(traffic 570 Generator) send the traffic to R1. There is no traffic from CE side. 572 Send X frames with X different SA and DA to DUT from R1 using traffic 573 generator. Stop the traffic at remote PE(R1).Measure the time taken 574 to remove these remote MACs from DUT PBB-EVPN MAC table. The DUT and 575 MHPE2 are running in SA mode. 577 Measurement : 579 Measure the time taken to flush the X remote MACs from DUT PBB-EVPN 580 MAC table due to aging Repeat these test and plot the data. The test 581 is repeated for "N" times and the values are collected. The aging is 582 calculated by averaging the values obtained by "N" samples. 584 Aging time for X Macs in sec = (T1+T2+..Tn/N) 586 3.13. How long it takes to learn both local and remote macs. 588 Objective: 590 To record the time taken to learn both local and remote macs. 592 Procedure: 594 Configure EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 595 running MPLS,BGP,RR is acting as route reflector to R1,MHPE2 and DUT. 596 Once the BGP is established. Record the DUT EVPN table. For MH PE 597 ESI must be configured per IFD/Interface. Using RT(traffic 598 generator) send the traffic to the routers. The traffic is bi 599 directional. 601 Send X frames with X different SA and DA to DUT from R1 using traffic 602 generator. Send X frames with different SA and DA from traffic 603 generator connected to CE. The SA and DA of flows must be 604 complimentary to have unicast flows. Measure the time taken by the 605 DUT to learn 2X in EVPN MAC. DUT and MHPE2 are running in SA mode. 607 Measurement : 609 Measure the time taken to learn 2X MAC address in DUT EVPN MAC table. 610 Repeat these test and plot the data. The test is repeated for "N" 611 times and the values are collected. The mac learning time is 612 calculated by averaging the values obtained by "N" samples. 614 Time to learn 2X Macs in sec = (T1+T2+..Tn/N) 616 3.14. PBB-EVPN How long it takes to learn both local and remote macs 618 Objective: 620 To record the time taken to learn both local and remote macs. 622 Procedure: 624 Configure PBB-EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 625 running MPLS,BGP,RR is acting as route reflector to R1,MHPE2 and DUT. 626 Once the BGP is established. Record the DUT EVPN table.For MH PE ESI 627 must be configured per IFD/Interface. Using RT(traffic generator) 628 send the traffic to the routers. 630 Send X frames with X different SA and DA to DUT from R1 using traffic 631 generator. Send X frames with different SA and DA from traffic 632 generator connected to CE. The SA and DA of flows must be 633 complimentary to have unicast flows. Measure the time taken by the 634 DUT to learn 2X in MAC table. DUT and MHPE2 are running in SA mode. 636 Measurement : 638 Measure the time taken to learn 2X MAC address table in DUT PBB-EVPN 639 MAC table. Repeat these test and plot the data. The test is 640 repeated for "N" times and the values are collected. The mac 641 learning time is calculated by averaging the values obtained by "N" 642 samples. 644 Time to learn 2X Macs in sec = (T1+T2+..Tn/N) 646 4. High Availability 648 4.1. To Record the whether there is traffic loss due to routing engine 649 failover for redundancy test. 651 Objective: 653 To record traffic loss during routing engine failover. 655 Procedure: 657 Configure 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 DUT. 659 Once the BGP is established. Record the DUT EVPN table. For MH PE 660 ESI must be configured per IFD/Interface. Using RT(traffic 661 generator) Send bi directional to the routers. 663 Send X frames from CE to DUT from traffic generator withX different 664 SA and DA. Send X frames from traffic generator to R1 with X 665 different SA and DA so that 2X MAC address will be learned in DUT. 666 There is a bi directional traffic flow with X pps in each direction. 667 Then do a routing engine fail-over. 669 Measurement : 671 There should be 0 traffic loss which is the ideal case, No change in 672 the DF role. DUT should not withdraw any routes.Repeat the test "N" 673 times and plot the data.The packet loss is calculated by averaging 674 the values obtained from "N" samples. 676 Packet loss in sec = (T1+T2+..Tn/N) 678 4.2. PBB-EVPN To Record the whether there is traffic loss due to 679 routing engine failover for redundancy test 681 Objective: 683 To record traffic loss during routing engine failover. 685 Procedure: 687 Configure PBB-EVPN EVI in R1,MHPE2,DUT. All 4 routers except CE are 688 running MPLS,BGP,RR is acting as route reflector to R1,MHPE2 and DUT. 689 Once the BGP is established. Record the DUT PBB-EVPN table. For MH 690 PE ESI must be configured per IFD/Interface. Using RT(traffic 691 generator) send the traffic to the routers. 693 Send X frames to DUT with X different SA and DA from CE using the 694 traffic generator. Send X frames from traffic generator to R1 with X 695 different SA and DA so that 2X MAC address will be Learned in DUT. 696 There is a bi directional traffic flow with X pps in each direction. 697 Then do a routing engine fail-over. 699 Measurement : 701 There should be 0 traffic loss which is the ideal case, No change in 702 the DF role. DUT should not withdraw any routes.Repeat the test "N" 703 times and plot the data.The packet loss is calculated by averaging 704 the values obtained from "N" samples. 706 Packet loss in sec = (T1+T2+..Tn/N) 708 5. ARP/ND Scale 710 These tests are conducted to Record the scaling parameter of ARP/ND 711 of the DUT. 713 5.1. To find ARP/ND scale 715 Objective: 717 To Record the ARP/ND scale of the DUT. 719 Procedure: 721 Configure EPN EVI in R1,MHPE2,DUT.All 4 routers except CE are running 722 MPLS,BGP,RR is acting as route reflector to R1,MHPE2 and DUT. Once 723 the BGP is established. Record the DUT EVPN table. For MH PE ESI 724 must be configured per IFD/Interface. Using RT(traffic generator) 725 send arp/ICMPv6 request to the DUT which has gateway configured. 727 Send X arp/icmpv6 request from RT to DUT with different sender ip/ 728 ipv6 address to the same target gateway ip address. Measure whether 729 X MAC+IPv4 address/MAC+IPv6 address of the hosts are learned in DUT. 731 Measurement : 733 The DUT must learn X MAC+IPV4/MAC+IPv6 and it must advertise the X 734 MAC+IPV4/MAC+IPV6 to the remote router. 736 6. Scale 738 This is to measure the performance of DUT in scaling to "X" EVPN 739 instances. The measured parameters are CPU usage, memory 740 leak,crashes. 742 6.1. To Measure the scale limit of DUT with trigger (Scale without 743 traffic) 745 Objective: 747 To measure the scale limit of DUT for EVPN. 749 Procedure: 751 Configure EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 752 running MPLS,BGP,RR is acting as route reflector to R1,MHPE2 and DUT. 753 Once the BGP is established. Record the DUT EVPN table. For 754 MHPE,DUT ESI must be configured per IFD/Interface. 756 The DUT,MHPE2 and R1 are scaled to "N" EVI.Clear BGP neighbors of the 757 DUT. Once adjacency is established in the DUT. Measure the routes 758 received from MHPE2 and R1 for "N" EVI in the DUT. 760 Measurement : 762 There should not be any loss of route types 1,2,3 and 4 in DUT. DUT 763 must relearn all type 1,2,3 and 4 from remote routers. The DUT must 764 be subjected to various values of N to find the optimal scale limit 766 6.2. PBB-EVPN To measure the scale limit with trigger. 768 Objective: 770 To measure the scale limit of DUT for PBB-EVPN. 772 Procedure: 774 Configure "N" PBB-EVPN instances in R1,MHPE2,DUT.All 4 routers except 775 CE are running MPLS,BGP,RR is acting route reflector to R1,MHPE2 and 776 DUT. Once BGP is established. Record the DUT PBB-EVPN table.For 777 MHPE ESI must be configured on IFD/Interface. 779 The DUT,MHPE2 and R1 are scaled to "N" PBB-EVPN instances. Clear BGP 780 neighbors in the DUT Once adjacency is established in DUT, check 781 routes received from R1 and MHPE2. 783 Measurement : 785 There should not be any loss of route types 2,3 and 4 in DUT. The 786 DUT must relearn all type 2,3 and 4 routes from remote routers. The 787 DUT must be subjected to various values of N to find the optimal 788 scale limit. 790 6.3. To measure the convergence time of DUT with scale and traffic. 792 Objective: 794 To measure the convergence time of DUT when the DUT is scaled with 795 EVPN instance along with traffic. 797 Procedure: 799 Configure EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 800 running MPLS,BGP,RR is acting as route reflector to R1,MHPE2 and DUT. 801 Once the BGP is established. Record the DUT EVPN table. For MH PE 802 ESI must be configured per IFD/Interface. Using RT(traffic 803 generator)send the traffic to the routers. 805 Scale N EVIs in DUT,R1 and MHPE2.Send F frames to DUT from CE using 806 traffic generator with X different SA and DA for N EVI's. Send F 807 frames from traffic generator to R1 with X different SA and DA. 808 There will be 2X number of MAC address will be learned in DUT EVPN 809 MAC table. There is a bi directional traffic flow with F pps in each 810 direction. Then clear the BGP neighbors in the DUT. Once the 811 adjacency is restored in DUT. Measure the time taken to learn 2X MAC 812 address in DUT MAC table. 814 Measurement : 816 The DUT must learn 2X MAC address. Measure the time taken to learn 817 2X MAC in DUT. Repeat these test and plot the data.The test is 818 repeated for "N" times and the values are collected.The convergence 819 time is calculated by averaging the values obtained by "N" samples. 821 Convergence time in sec = (T1+T2+..Tn/N) 823 6.4. .PBB-EVPN To measure the convergence time of DUT with scale and 824 traffic. 826 Objective: 828 To measure the convergence time of DUT when the DUT is scaled with 829 PBB-EVPN instance along with traffic. 831 Procedure: 833 Configure PBB-EVPN instances in R1,MHPE2,DUT.All 4 routers except CE 834 are running MPLS,BGP,RR is acting as route reflector to R1,MHPE2 and 835 DUT. Once BGP is established. Record the DUT EVPN table. For MH PE 836 ESI must be configured per IFD/Interface. Using RT(traffic 837 generator) send the traffic to the routers. 839 Scale N PBB-EVI's in DUT,R1 and MHPE2.Send F frames to DUT from CE 840 using traffic generator with X different SA and DA for N EVI's. Send 841 F frames from traffic generator to R1 with X different SA and DA. 842 There will be 2X number of MAC address will be learned in DUT PBB- 843 EVPN MAC table. There is a bi directional traffic flow with F pps in 844 each direction. Then clear the BGP neighbors in the DUT. Once the 845 adjacency is restored in DUT. Measure the time taken to learn 2X MAC 846 address in DUT PBB-MAC table. 848 Measurement : 850 The DUT must learn 2X MAC address. Measure the time taken to learn 851 2X MAC in DUT. Repeat these test and plot the data. The test is 852 repeated for "N" times and the values are collected. The convergence 853 time is calculated by averaging the values obtained by "N" samples. 855 Convergence time in sec = (T1+T2+..Tn/N) 857 7. SOAK Test 859 This is measuring the performance of DUT running with scaled 860 configuration with traffic over a peroid of time "T'". In each 861 interval "t1" the parameters measured are CPU usage, memory usage, 862 crashes. 864 7.1. To Measure the stability of the DUT with scale and traffic. 866 Objective: 868 To measure the stability of the DUT in a scaled environment with 869 traffic. 871 Procedure: 873 Configure EVPN EVI in R1,MHPE2,DUT.All 4 routers except CE are 874 running MPLS,BGP,RR is acting as route reflector to R1,MHPE2 and DUT. 875 Once the BGP is established. Record the DUT EVPN table. For MH PE 876 ESI must be configured per IFD/Interface. Using RT(traffic 877 generator) send the traffic to the routers. 879 Scale N EVI's in DUT,R1 and MHPE2.Send F frames to DUT from CE using 880 traffic generator with different X SA and DA for N EVI's. Send F 881 frames from traffic generator to R1 with X different SA and DA. 882 There will be 2X number of MAC address will be learned in DUT EVPN 883 MAC table. There is a bi directional traffic flow with F pps in each 884 direction. The DUT must run with traffic for 24 hours, every hour 885 check for memory leak, crash. 887 Measurement : 889 Take the hourly reading of CPU, process memory. There should not be 890 any leak, crashes, CPU spikes. 892 7.2. PBB-EVPN to measure the stability of DUT with scale and traffic. 894 Objective: 896 To measure the stability of the DUT in a scaled environment with 897 traffic. 899 Procedure: 901 Configure N PBB-EVPN instances in R1, MHPE2, DUT. All 4 routers 902 except CE are running MPLS,BGP,RR is acting as route reflector to 903 R1,MHPE2 and DUT. Once the BGP comes up Record the DUT EVPN table. 904 for MH PE ESI must be configured per IFD/Interface. Using RT(traffic 905 generator)send the traffic to the routers. 907 Scale N PBB-EVI's in DUT,R1 and MHPE2.Send F frames to DUT from CE 908 using traffic generator with X different SA and DA for N EVI's. Send 909 F frames from traffic generator to R1 with X different SA and DA. 910 There will be 2X number of MAC address will be learned in DUT PBB- 911 EVPN MAC table. There is a bi directional traffic flow with F pps in 912 Each direction. The DUT must run with traffic for 24 hours, every 913 hour check the memory leak, crashes. 915 Measurement : 917 Take the hourly reading of CPU process, memory usages. There should 918 not be any memory leak, crashes,CPU spikes. 920 8. Acknowledgements 922 We would like to thank Fioccola Giuseppe of Telecom Italia reviewing 923 our draft and commenting it. We would like to thank Sarah Banks for 924 guiding and mentoring us. 926 9. IANA Considerations 928 This memo includes no request to IANA. 930 10. Security Considerations 932 There is no additional consideration from RFC 6192. 934 11. References 936 11.1. Normative References 938 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 939 Requirement Levels", BCP 14, RFC 2119, 940 DOI 10.17487/RFC2119, March 1997, 941 . 943 [RFC2544] Bradner, S. and J. McQuaid, "Benchmarking Methodology for 944 Network Interconnect Devices", RFC 2544, 945 DOI 10.17487/RFC2544, March 1999, 946 . 948 [RFC2899] Ginoza, S., "Request for Comments Summary RFC Numbers 949 2800-2899", RFC 2899, DOI 10.17487/RFC2899, May 2001, 950 . 952 11.2. Informative References 954 [RFC7432] Sajassi, A., Ed., Aggarwal, R., Bitar, N., Isaac, A., 955 Uttaro, J., Drake, J., and W. Henderickx, "BGP MPLS-Based 956 Ethernet VPN", RFC 7432, DOI 10.17487/RFC7432, February 957 2015, . 959 [RFC7623] Sajassi, A., Ed., Salam, S., Bitar, N., Isaac, A., and W. 960 Henderickx, "Provider Backbone Bridging Combined with 961 Ethernet VPN (PBB-EVPN)", RFC 7623, DOI 10.17487/RFC7623, 962 September 2015, . 964 Appendix A. Appendix 966 Authors' Addresses 968 Sudhin Jacob (editor) 969 Juniper Networks 970 Bangalore 971 India 973 Phone: +91 8061212543 974 Email: sjacob@juniper.net 976 Kishore Tiruveedhula 977 Juniper Networks 978 10 Technology Park Dr 979 Westford, MA 01886 980 USA 982 Phone: +1 9785898861 983 Email: kishoret@juniper.net