idnits 2.17.1 draft-balarajah-bmwg-ngfw-performance-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 (December 7, 2017) is 2324 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- No issues found here. Summary: 0 errors (**), 0 flaws (~~), 1 warning (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Benchmarking Methodology Working Group BB. Balarajah 3 Internet-Draft EANTC AG 4 Intended status: Informational December 7, 2017 5 Expires: June 10, 2018 7 Benchmarking Methodology for Network Security Device Performance 8 draft-balarajah-bmwg-ngfw-performance-00 10 Abstract 12 This document provides benchmarking terminology and methodology for 13 next-generation network security devices including next-generation 14 firewalls (NGFW), intrusion detection and prevention solutions (IDS/ 15 IPS) and unified threat management (UTM) implementations. The 16 document aims to strongly improve the applicability, reproducibility 17 and transparency of benchmarks and to align the test methodology with 18 today's increasingly complex 7application use cases. The main areas 19 covered in this document are test terminology, traffic profiles and 20 benchmarking methodology for NGFWs to start with. 22 Status of This Memo 24 This Internet-Draft is submitted in full conformance with the 25 provisions of BCP 78 and BCP 79. 27 Internet-Drafts are working documents of the Internet Engineering 28 Task Force (IETF). Note that other groups may also distribute 29 working documents as Internet-Drafts. The list of current Internet- 30 Drafts is at https://datatracker.ietf.org/drafts/current/. 32 Internet-Drafts are draft documents valid for a maximum of six months 33 and may be updated, replaced, or obsoleted by other documents at any 34 time. It is inappropriate to use Internet-Drafts as reference 35 material or to cite them other than as "work in progress." 37 This Internet-Draft will expire on June 10, 2018. 39 Copyright Notice 41 Copyright (c) 2017 IETF Trust and the persons identified as the 42 document authors. All rights reserved. 44 This document is subject to BCP 78 and the IETF Trust's Legal 45 Provisions Relating to IETF Documents 46 (https://trustee.ietf.org/license-info) in effect on the date of 47 publication of this document. Please review these documents 48 carefully, as they describe your rights and restrictions with respect 49 to this document. Code Components extracted from this document must 50 include Simplified BSD License text as described in Section 4.e of 51 the Trust Legal Provisions and are provided without warranty as 52 described in the Simplified BSD License. 54 Table of Contents 56 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 57 2. Requirements . . . . . . . . . . . . . . . . . . . . . . . . 2 58 3. Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 59 4. Test Setup . . . . . . . . . . . . . . . . . . . . . . . . . 3 60 4.1. Testbed Configuration . . . . . . . . . . . . . . . . . . 3 61 4.2. DUT/SUT Configuration . . . . . . . . . . . . . . . . . . 4 62 4.3. Test Equipment Configuration . . . . . . . . . . . . . . 6 63 4.3.1. Client Configuration . . . . . . . . . . . . . . . . 7 64 4.3.2. Backend Server Configuration . . . . . . . . . . . . 8 65 4.3.3. Traffic Flow Definition . . . . . . . . . . . . . . . 9 66 4.3.4. Traffic Load Profile . . . . . . . . . . . . . . . . 10 67 5. Test Bed Considerations . . . . . . . . . . . . . . . . . . . 11 68 6. Reporting . . . . . . . . . . . . . . . . . . . . . . . . . . 12 69 6.1. Key Performance Indicators . . . . . . . . . . . . . . . 13 70 7. Benchmarking Tests . . . . . . . . . . . . . . . . . . . . . 14 71 7.1. Throughput Performance . . . . . . . . . . . . . . . . . 15 72 7.1.1. Objective . . . . . . . . . . . . . . . . . . . . . . 15 73 7.1.2. Test Setup . . . . . . . . . . . . . . . . . . . . . 15 74 7.1.3. Test Parameters . . . . . . . . . . . . . . . . . . . 15 75 7.1.4. Test Procedures and expected Results . . . . . . . . 17 76 7.2. TCP Concurrent Connection Capacity . . . . . . . . . . . 18 77 7.3. TCP Connection Setup Rate . . . . . . . . . . . . . . . . 18 78 7.4. Application Transaction Rate . . . . . . . . . . . . . . 18 79 7.5. SSL/TLS Handshake Rate . . . . . . . . . . . . . . . . . 18 80 8. Formal Syntax . . . . . . . . . . . . . . . . . . . . . . . . 18 81 9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 18 82 10. Security Considerations . . . . . . . . . . . . . . . . . . . 18 83 11. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 18 84 12. Normative References . . . . . . . . . . . . . . . . . . . . 18 85 Appendix A. An Appendix . . . . . . . . . . . . . . . . . . . . 18 86 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 19 88 1. Introduction 90 TBD 92 2. Requirements 94 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 95 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 96 document are to be interpreted as described in RFC 2119 [RFC2119]. 98 3. Scope 100 TBD. 102 4. Test Setup 104 Test setup defined in this document will be applicable to all of the 105 benchmarking test cases described in Section 7 (Section 7). 107 4.1. Testbed Configuration 109 Testbed configuration MUST ensure that any performance implications 110 that are discovered during the benchmark testing aren't due to the 111 inherent physical network limitations such as number of physical 112 links and forwarding performance capabilities (throughput and 113 latency) of the network devise in the testbed. For this reason, this 114 document recommends to avoid external devices such as switch and 115 router in the testbed as possible. 117 In the typical deployment, the security devices (DUT/SUT) will not 118 have a large number of entries in MAC or ARP tables, which impact the 119 actual DUT/SUT performance due to MAC and ARP table lookup processes. 120 Therefore, depend on number of used IP address in client and server 121 side, it is recommended to connect Layer 3 device(s) between test 122 equipment and DUT/SUT as shown in figure 1 (Figure 1). 124 If the test equipment is capable to emulate layer 3 routing 125 functionality and there is no need for test equipment ports 126 aggregation, it is recommended to configure the test setup as shown 127 in figure 2 (Figure 2). 129 +-------------------+ +-----------+ +--------------------+ 130 |Aggregation Switch/| | | | Aggregation Switch/| 131 | Router +------+ DUT/SUT +------+ Router | 132 | | | | | | 133 +----------+--------+ +-----------+ +----------+---------+ 134 | | 135 | | 136 +-----------+-----------+ +------------+----------+ 137 | | | | 138 | +-------------------+ | | +-------------------+ | 139 | | Emulated Router(s)| | | | Emulated Router(s)| | 140 | | (Optional) | | | | (Optional) | | 141 | +-------------------+ | | +-------------------+ | 142 | +-------------------+ | | +-------------------+ | 143 | | Clients | | | | Servers | | 144 | +-------------------+ | | +-------------------+ | 145 | | | | 146 | Test Equipment | | Test Equipment | 147 +-----------------------+ +-----------------------+ 149 Figure 1: Testbed Setup - Option 1 151 +-----------------------+ +-----------------------+ 152 | +-------------------+ | +-----------+ | +-------------------+ | 153 | | Emulated Router(s)| | | | | | Emulated Router(s)| | 154 | | (Optional) | +----- DUT/SUT +-----+ (Optional) | | 155 | +-------------------+ | | | | +-------------------+ | 156 | +-------------------+ | +-----------+ | +-------------------+ | 157 | | Clients | | | | Servers | | 158 | +-------------------+ | | +-------------------+ | 159 | | | | 160 | Test Equipment | | Test Equipment | 161 +-----------------------+ +-----------------------+ 163 Figure 2: Testbed Setup - Option 2 165 4.2. DUT/SUT Configuration 167 An unique DUT/SUT configuration MUST be used for all of the 168 benchmarking tests described in section 7 (Section 7). Since each 169 DUT/SUT will have their own unique configuration, users SHOULD 170 configure their device with the same parameters that would be used in 171 the actual deployment of the device or a typical deployment. Also it 172 is mandatory to enable all the security features on the DUT/SUT in 173 order to achieve maximum security coverage for a specific deployment 174 scenario. 176 This document attempts to define the recommended security features 177 which SHOULD be consistently enabled for all test cases. The table 178 below describes the recommended sets of feature list which SHOULD be 179 configured on the DUT/SUT. In order to improve repeatability, a 180 summary of the DUT configuration including description of all enabled 181 DUT/SUT features MUST be published with the benchmarking results. 183 +----------------------------------------------------+ 184 | Device | 185 +---------------------------------+--+----+---+------+ 186 | | | | | | SSL | 187 | NGFW |NGIPS|AD| WAF|BPS|Broker| 188 +----------------------------------------------------------------------+ 189 | | |Included |Added to| Future test standards | 190 | DUT Features |Feature|in initial|future | to be de^eloped | 191 | | |Scope |Scope | | 192 +---------------------------------------------------+---+---+---+------+ 193 | SSL Inspection | x | | x | | | | | | 194 +----------------------------------------------------------------------+ 195 | IDS/IPS | x | x | | | | | | | 196 +----------------------------------------------------------------------+ 197 | Web Filtering | x | | x | | | | | | 198 +----------------------------------------------------------------------+ 199 | Anti^irus | x | x | | | | | | | 200 +----------------------------------------------------------------------+ 201 | Anti Spyware | x | x | | | | | | | 202 +----------------------------------------------------------------------+ 203 | Anti Botnet | x | x | | | | | | | 204 +----------------------------------------------------------------------+ 205 | DLP | x | | x | | | | | | 206 +----------------------------------------------------------------------+ 207 | DDoS | x | | x | | | | | | 208 +----------------------------------------------------------------------+ 209 | SSL Certificate | x | | x | | | | | | 210 | Validation | | | | | | | | | 211 +----------------------------------------------------------------------+ 212 | Logging and | x | x | | | | | | | 213 | Reporting | | | | | | | | | 214 +----------------------------------------------------------------------+ 215 | Application | x | x | | | | | | | 216 | Identification | | | | | | | | | 217 +-----------------+-------+----------+--------+-----+---+---+---+------+ 219 Table 1: DUT/SUT Feature List 221 It is also recommended to configure a realistic number of access 222 policy rules on the DUT/SUT. This document attempts to determine the 223 number of access policy rules for three different class of DUT/SUT. 225 The document classified the DUT/SUT based on its performance 226 capability. The access rule defined in the, MUST be configured from 227 top to bottom in correct order. The configured access policy rule 228 MUST NOT block the test traffic used for the performance test. 230 +---------------------------------------------------+------------------+ 231 | | DUT/SUT | 232 | | Classification | 233 | | # Rules | 234 +-----------+-----------+--------------------+------+------------------+ 235 | | Match | | | 236 |Rules Type | Criteria| Description |Action|Small|Medium|Large| 237 +----------------------------------------------------------------------+ 238 |Application|Application|Any application |block | 10 | 20 | 50 | 239 |layer | |traffic NOT included| | | | | 240 | | |in the test traffic | | | | | 241 +----------------------------------------------------------------------+ 242 |Transport |Src IP and |Any src IP used in |block | 50 | 100 | 250 | 243 |layer |TCP/UDP |the test AND any dst| | | | | 244 | |Dst ports |ports NOT used in | | | | | 245 | | |the test traffic | | | | | 246 +----------------------------------------------------------------------+ 247 |IP layer |Src/Dst IP |Any src/dst IP NOT |block | 50 | 100 | 250 | 248 | | |used in the test | | | | | 249 +----------------------------------------------------------------------+ 250 |Application|Application|Applications |allow | 10 | 10 | 10 | 251 |layer | |included in the test| | | | | 252 | | |traffic | | | | | 253 +----------------------------------------------------------------------+ 254 |Transport |Src IP and |Half of the src IP |allow | 1 | 1 | 1 | 255 |layer |TCP/UDP |used in the test AND| | | | | 256 | |Dst ports |any dst ports used | | | | | 257 | | |in the test traffic.| | | | | 258 | | |One rule per subnet | | | | | 259 +----------------------------------------------------------------------+ 260 |IP layer |Src IP |The rest of the src |allow | 1 | 1 | 1 | 261 | | |IP subnet range used| | | | | 262 | | |in the test. | | | | | 263 | | |One rule per subnet | | | | | 264 +-----------+--------------------------------+------+-----+------+-----+ 266 Table 2: DUT/SUT Access List 268 4.3. Test Equipment Configuration 270 In general, test equipment allows configuring parameters in different 271 protocol level. These parameters thereby influencing the traffic 272 flows which will be offered and impacting performance measurements. 274 This document attempts to explicitly specify which test equipment 275 parameters SHOULD be configurable, any such parameter(s) MUST be 276 noted in the test report. 278 4.3.1. Client Configuration 280 This section specifies which parameters SHOULD be considerable while 281 configuring emulated clients using test equipment. Also this section 282 specifies the recommended values for certain parameters. 284 4.3.1.1. TCP Stack Attributes 286 The TCP stack SHOULD use a TCP Reno variant, which include congestion 287 avoidance, back off and windowing, retransmission and recovery on 288 every TCP connection between client and server endpoints. The 289 default IPv4 and IPv6 MSS segments size MUST be set to 1460 bytes and 290 1440 bytes and a TX and RX receive windows of 32768 bytes. Delayed 291 ACKs are permitted, but it SHOULD be limited to either a 200 mSec 292 delay timeout or 3000 in bytes before a forced ACK. Up to 3 retries 293 SHOULD be allowed before a timeout event is declared. All traffic 294 MUST set the TCP PSH flag to high. The source port range SHOULD be 295 in the range of 1024 - 65535. Internal timeout SHOULD be dynamically 296 scalable per RFC 793.. 298 4.3.1.2. Client IP Address Space 300 The sum of the client IP space SHOULD contain the following 301 attributes. The traffic blocks SHOULD consist of multiple unique, 302 continuous static address blocks. A default gateway is permitted. 303 The IPv4 ToS byte should be set to '00'. 305 The following equation can be used to determine the required total 306 number of client IP address. 308 Desired total number of client IP = Target throughput [Mbit/s] / 309 Throughput per IP address [Mbit/s] 311 (Idea 1) 6-7 Mbps per IP= 1,400-1,700 IPs per 10Gbit/s throughput 313 (Idea 2) 0.1-0.2 Mbps per IP = 50,000-100,000 IPs per 10Gbit/s 314 throughput 316 Based on deployment and usecase scenario, client IP addresses SHOULD 317 be distributed between IPv4 and IPv6 type. This document recommends 318 using the following ratio(s) between IPv4 and IPv6: 320 (Idea 1) 100 % IPv4, no IPv6 321 (Idea 2) 80 % IPv4, 20 % IPv6 323 (Idea 3) 50 % IPv4, 50 % IPv6 325 (Idea 4) 0 % IPv4, 100 % IPv6 327 4.3.1.3. Emulated Web Browser Attributes 329 The emulated web browser contains attributes that will materially 330 affect how traffic is loaded. The objective is to emulate a modern, 331 typical browser attributes to improve realism of the result set. The 332 emulated browser must negotiate HTTP 1.1 with persistence. The 333 browser will open up to 6 TCP connections per Server endpoint IP at 334 any time depending on how many sequential transactions are needed to 335 be processed. Within the TCP connection multiple transactions can be 336 processed if the emulated browser has available connections, for 337 example where transactions to the same server endpoint IP exceed 6 or 338 are non-sequential. The browser must advertise a User-Agent header. 339 Headers will be sent uncompressed. The browser should enforce 340 content length validation. 342 4.3.1.4. Client Emulated Web Browser SSL/TLS Layer Attributes 344 The test traffic shall be a realistic blend of encrypted and clear 345 traffic. For encrypted traffic, the following attributes shall 346 define the negotiated encryption parameters. The tests must use 347 TLSv1.2 or higher with a record size of 16383, commonly used cipher 348 suite and key strength. Session reuse or ticket resumption may be 349 used for subsequent connections to the same Server endpoint IP. The 350 client endpoint must send TLS Extension SNI information when opening 351 up a security tunnel. Server certificate validation should be 352 disabled. 354 If the DUT/SUT doesn't perform SSL inspection, cipher suite and 355 certificate selection for the test is irrelevant. However, it is 356 recommended to use latest and not deprecated certificates, in order 357 to mimic real world traffic. 359 4.3.2. Backend Server Configuration 361 This document attempts to specify which parameters should be 362 considerable while configuring emulated backend servers using test 363 equipment. 365 4.3.2.1. TCP Stack Attributes 367 The TCP stack SHOULD use a TCP Reno variant, which include congestion 368 avoidance, back off and windowing, retransmission and recovery on 369 every TCP connection between client and server endpoints. The 370 default IPv4 MSS segment size MUST be set to 1460 bytes and a TX and 371 RX receive windows of at least 32768 bytes. Delayed ACKs are 372 permitted but SHOULD be limited to either a 200 mSec delay timeout or 373 3k in bytes before a forced ACK. Up to 2 retries SHOULD be allowed 374 before a timeout event is declared. All traffic must set the TCP PSH 375 flag to high. The source port range SHOULD be in the range of 1024 - 376 65535. Internal timeout should be dynamically scalable per RFC 793. 378 4.3.2.2. Server Endpoint IP Addressing 380 The server IP blocks should consist of unique, continuous static 381 address blocks with one IP per Server FQDN endpoint per test port. 382 The IPv4 ToS byte should be set to '00'. The source mac address of 383 the server endpoints shall be the same emulating routed behavior. 384 Each Server FQDN should have it's own unique IP address. The Server 385 IP addressing should be fixed to the same number of FQDN entries. 387 4.3.2.3. HTTP / HTTPS Server Pool Endpoint Attributes 389 The emulated server pool for HTTP should listen on TCP port 80 and 390 emulated HTTP version 1.1 with persistence. For HTTPS server, the 391 pool must have the same basic attributes of an HTTP server pool plus 392 attributes for SSL/TLS. The server must advertise a server type. 393 For HTTPS server, TLS 1.2 or higher must be used with a record size 394 of 16,383 bytes and ticket resumption or Session ID reuse enabled. 395 The server must listen on port TCP 443. The server shall serve a 396 2048 server SSL certificate to the client. It is required that the 397 HTTPS server also check Host SNI information with the Fully Qualified 398 Domain Name (FQDN). Client certificate validation should be 399 disabled. 401 If the DUT/SUT doesn't perform SSL inspection, cipher suite and 402 certificate selection for the test is irrelevant. However, it is 403 recommended to use latest and not deprecated certificates, in order 404 to mimic real world traffic. 406 4.3.3. Traffic Flow Definition 408 The section describes the traffic pattern between the client and 409 server endpoints. At the beginning of the test, the server endpoint 410 initializes and will be in a ready to accept connection state 411 including initialization of the TCP stack as well as bound HTTP and 412 HTTPS servers. When a client endpoint is needed, it will initialize 413 and be given attributes such as the MAC and IP address. The behavior 414 of the client is to sweep though the given server IP space, 415 sequentially generating a recognizable service by the DUT. Thus, a 416 balanced, mesh between client endpoints and server endpoints will be 417 generated in a client port server port combination. Each client 418 endpoint performs the same actions as other endpoints, with the 419 difference being the source IP of the client endpoint and the target 420 server IP pool. The client shall use Fully Qualified Domain Names in 421 Host Headers and for TLS 1.2 Server Name Indication (SNI). 423 4.3.3.1. Description of Intra-Client Behavior 425 Client endpoints are independent of other clients that are 426 concurrently executing. When a client endpoint initiate traffic, 427 this section will describe how the steps though different services. 428 Once initialized, the user should randomly hold (perform no 429 operation) for a few milliseconds to allow for better randomization 430 of start of client traffic. The client will then either open up a 431 new TCP connection or connect to a TCP persistence stack still open 432 to that specific server. At any point that the service profile may 433 require encryption, a TLS 1.2 encryption tunnel will form presenting 434 the URL request to the server. The server will then perform an SNI 435 name check with the proposed FQDN compared to the domain embedded in 436 the certificate. Only when correct, will the server process the 437 object. The initial object to the server does not have a fixed size, 438 its size is based on for example the URL path length. Up to six 439 additional sub-URLs (Objects on the service page) may be requested 440 simultaneously. This may or may not be to the same server IP as the 441 initial URL. Each sub-object will also use a conical FQDN and URL 442 path, as observed in the traffic mix used. The traffic mix in the 443 appendix table is represented by the actions of each and every client 444 endpoint. Therefor the instantaneous percent of mix will vary, but 445 the overall mix through the duration of the test will be fixed. This 446 is based on the number of active users, TCP recovery mechanism, etc. 448 4.3.4. Traffic Load Profile 450 The loading of traffic will be described in this section. The 451 loading of an traffic load profile has five distinct phases: Init, 452 ramp up, sustain, ramp down/close, and collection. 454 Within the Init phase, test bed devices including the client and 455 server endpoints should negotiate layer 2-3 connectivity such as MAC 456 learning and ARP. Only after successful MAC learning or ARP 457 resolution shall the test iteration move to the next phase. No 458 measurements are made in this phase. The minimum recommended time 459 for init phase is 5 seconds. During this phase the emulated clients 460 SHOULD NOT initiate any sessions with the DUT/SUT, in contrast, the 461 emulated servers should be ready to accept requests from DUT/SUT or 462 from emulated clients. 464 In the ramp up phase, the test equipment should start to generate the 465 test traffic. It should use a set approximate number of unique 466 client IP addresses actively to generate traffic. The traffic should 467 ramp from zero to desired target throughput objective. The duration 468 for the ramp up phase must be configured long enough, so that the 469 test equipment does not overwhelm DUT/SUT's supported performance 470 metrics, namely: connection setup rate, concurrent connection and 471 application transaction. The recommended time duration for the ramp 472 up phase is 180-300 seconds. No measurements are made in this phase. 474 In the sustain phase, the test equipment should keep to generate 475 traffic at constant rate for a constant number of active client IPs. 476 The recommended time duration for sustain phase is 600 seconds. This 477 is the phase where measurements occur. 479 In the ramp down/close phase, no new connection is established and no 480 measurements are made. The recommend duration of this phase is 180- 481 300 seconds. 483 The last phase is administrative and will be when the tester merges 484 and collates the report data. 486 5. Test Bed Considerations 488 This section recommends steps to control the test environment and 489 test equipment, specifically focusing on virtualized environments and 490 virtualized test equipment. 492 1. Ensure that any ancillary switching or routing functions between 493 the system under test and the test equipment do not limit the 494 performance of the traffic generator. This is specifically 495 important for virtualized components (vSwitches, vRouters). 497 2. Verify that the performance of the test equipment matches and 498 reasonably exceeds the expected maximum performance of the system 499 under test. 501 3. Assert that the test bed characteristics are stable during the 502 whole test session. A number of factors might influence 503 stability specifically for virtualized test beds, for example 504 additional work loads in a virtualized system, load balancing and 505 movement of virtual machines during the test, or simple issues 506 such as additional heat created by high workloads leading to an 507 emergency CPU performance reduction. 509 Test bed reference pre-tests help to ensure that the desired traffic 510 generator aspects such as maximum throughput and the network 511 performance metrics such as maximum latency and maximum packet loss 512 are met. 514 Once the desired maximum performance goals for the system under test 515 have been identified, a safety margin of 10 % SHOULD be added for 516 throughput and subtracted for maximum latency and maximum packet 517 loss. 519 Test bed preparation can be performed either by configuring the DUT 520 in the most trivial setup (fast forwarding) or without presence of 521 DUT. 523 6. Reporting 525 This section describes how the final report should be formatted and 526 presented. The final test report may have two major sections; 527 Introduction and result sections. The following attributes should be 528 present in the introduction section of the test report. 530 1. The name of the NetSecOPEN traffic mix must be prominent. 532 2. The time and date of the execution of the test must be prominent. 534 3. Summary of testbed software and Hardware details 536 A. DUT Hardware/Virtual Configuration 538 + This section should clearly identify the make and model of 539 the DUT 541 + iThe port interfaces, including speed and link information 542 must be documented. 544 + If the DUT is a virtual VNF, interface acceleration such 545 as DPDK and SR-IOV must be documented as well as cores 546 used, RAM used, and the pinning / resource sharing 547 configuration. The Hypervisor and version must be 548 documented. 550 + Any additional hardware relevant to the DUT such as 551 controllers must be documented 553 B. DUT Software 555 + The operating system name must be documented 556 + The version must be documented 558 + The specific configuration must be documented 560 C. DUT Enabled Features 562 + Specific features, such as logging, NGFW, DPI must be 563 documented 565 + iAttributes of those featured must be documented 567 + Any additional relevant information about features must be 568 documented 570 D. Test equipment hardware and software 572 + Test equipment vendor name 574 + Hardware details including model number, interface type 576 + Test equipment firmware and test application software 577 version 579 4. Results Summary / Executive Summary 581 1. Results should resemble a pyramid in how it is reported, with 582 the introduction section documenting the summary of results 583 in a prominent, easy to read block. 585 2. In the result section of the test report, the following 586 attributes should be present for each test scenario. 588 a. KPIs must be documented separately for each test 589 scenario. The format of the KPI metrics should be 590 presented as described in section 6.1 (Section 6.1). 592 b. The next level of detains should be graphs showing each 593 of these metrics over the duration (sustain phase) of the 594 test. This allows the user to see the measured 595 performance stability changes over time. 597 6.1. Key Performance Indicators 599 This section lists KPIs for overall benchmarking tests scenarios. 600 All KPIs MUST be measured in whole period of sustain phase as 601 described insection 4.3.4 (Section 4.3.4). All KPIs MUST be measured 602 from test equipment statistics only. 604 o TCP Concurrent Connection Capacity 605 This key performance indicator will measure the average concurrent 606 open TCP connections in the sustaining period. 608 o TCP Connection Setup Rate 609 This key performance indicator will measure the average 610 established TCP connections per second in the sustaining period. 611 For Session setup rate benchmarking test scenario, the KPI will 612 measure average established and terminated TCP connections per 613 second simultaneously. 615 o Application Transaction Rate 616 This key performance indicator will measure the average successful 617 transactions per seconds in the sustaining period. 619 o TLS Handshake Rate 620 This key performance indicator will measure the average TLS 1.2 or 621 higher session formation rate within the sustaining period. 623 o URL Response time / Time to Last Byte (TTLB) 624 This key performance indicator will measure the minimum, average 625 and maximum per URL response time in the sustaining period as well 626 as the average variance in the same period. 628 o Application Transaction Time 629 This key performance indicator will measure the minimum, average 630 and maximum the amount of time to receive all objects from the 631 server. 633 o Time to First Byte (TTFB) 634 This key performance indicator will measure minimum, average and 635 maximum the time to first byte. TTFB is the elapsed time between 636 sending the SYN packet from the client and receiving the first 637 byte of application date from the DUT/SUT. TTFB SHOULD be 638 expressed in millisecond. 640 o TCP Connect Time 641 This key performance indicator will measure minimum, average and 642 maximum TCP connect time. It is elapsed between the time the 643 client sends a SYN packet and the time it receives the SYN/ACK. 644 TCP connect time SHOULD be expressed in millisecond. 646 7. Benchmarking Tests 647 7.1. Throughput Performance 649 7.1.1. Objective 651 To determine the average throughput performance of the DUT/SUT when 652 using application traffic mix defined insection 7.1.3.3 653 (Section 7.1.3.3). 655 7.1.2. Test Setup 657 Test bed setup MUST be configured as defined in section 4 658 (Section 4). Any test scenario specific test bed configuration 659 changes must be documented. 661 7.1.3. Test Parameters 663 In this section, test scenario specific parameters SHOULD be defined. 665 7.1.3.1. Test Equipment Configuration Parameters 667 Test equipment configuration parameters MUST conform to the 668 requirements defined in section 4.3 (Section 4.3). Following 669 parameters MUST be noted for this test scenario: 671 Client IP address range 673 Server IP address range 675 Traffic distribution ratio between IPv4 and IPv6 677 Traffic load objective or specification type (e.g Throughput, 678 SimUsers and etc.) 680 Target throughput: It can be defined based on requirements. 681 Otherwise it represents aggregated line rate of interface(s) used 682 in the DUT/SUT 684 Initial throughput: Initial throughput can be up to 10% of the 685 "Target throughput" 687 7.1.3.2. DUT/SUT Configuration Parameters 689 DUT/SUT parameters MUST conform to the requirements defined in 690 section 4.2 (Section 4.2). Any configuration changes for this 691 specific test scenario MUST be documented. 693 7.1.3.3. Traffic Profile 695 Test scenario MUST be run with a single application traffic mix 696 profile. The name of the NetSecOpen traffic mix MUST be documented. 698 7.1.3.4. Test Results Acceptance Criteria 700 The following test Criteria is defined as test results acceptance 701 criteria 703 a. Number of failed Application transaction MUST be 0.01%. 705 b. Number of Terminated TCP connection due to unexpected TCP RST 706 sent by DUT/SUT MUST be less than 0.01% 708 c. Maximum deviation (max. dev) of application transaction time / 709 TTLB (Time To Last Byte) MUST be less than X (e.g. 2, TBD) 710 The following equation MUST be used to calculate the deviation of 711 application transaction time or TTLB. 713 max. dev = max((avg_latency - min_latency),(max_latency - 714 avg_latency)) / (Initial latency) 716 Where, the initial latency is calculated using the following 717 equation. For this calculation, the latency values (min', avg' 718 and max') MUST be measured during test procedure step 1 as 719 defined in section 7.1.4.1 (Section 7.1.4.1). 720 The variable latency represents application transaction time or 721 TTLB. 723 Initial latency:= min((avg' latency - min' latency) | (max' 724 latency - avg' latency)) 726 d. Maximum value of TCP connect time must be less than (TBD) ms. 727 (beta tests required to determine the value). The definition for 728 TCP connect time can be found in section 6.2 (Section 6.1). 730 e. Maximum value of Time to First Byte must be less than 2* TCP 731 connect time. 733 Test Acceptance criteria for this test scenario MUST be monitored 734 during the sustain phase of the traffic load profile only. 736 7.1.3.5. Measurement 738 Following KPI metrics MUST be reported for this test scenario. 740 Mandatory KPIs: average Throughput, maximum Concurrent TCP 741 connection, TTLB/application transaction time (minimum, average and 742 maximum) and average application transaction rate 744 Optional KPIs: average TCP connection setup rate, average TLS 745 handshake rate, TCP connect time and TTFB 747 7.1.4. Test Procedures and expected Results 749 The test procedure is designed to measure the throughput performance 750 of the DUT/SUT at the sustaining period of traffic load profile. The 751 test procedure consists of three major steps. 753 7.1.4.1. Step 1: Test Initialization and Qualification 755 Verify the link status of the all connected physical interfaces. All 756 interfaces are expected to be "UP" status. 758 Configure traffic load profile of the test equipment to generate test 759 traffic at "initial throughput" rate as described in the parameters 760 section. The DUT/SUT SHOULD reach the "initial throughput" during 761 the sustain phase. Measure all KPI as defined in section 7.1.3.5 762 (Section 7.1.3.5). The measured KPIs during the sustain phase MUST 763 meet acceptance criteria "a" and "b" defined in section 7.1.3.4 764 (Section 7.1.3.4). 766 If the KPI metrics do not meet the acceptance criteria, the test 767 procedure MUST NOT be continued to step 2. 769 7.1.4.2. Step 2: Test Run with Target Objective 771 Configure test equipment to generate traffic at "Target throughput" 772 rate defined in the parameter table. The test equipment SHOULD 773 follow the traffic load profile definition as described in section 774 4.3.4 (Section 4.3.4). The test equipment SHOULD start to measure 775 and record all specified KPIs. The frequency of KPI metrics 776 measurement MUST be less than 5 seconds. Continue the test until all 777 traffic profile phases are completed. 779 The DUT/SUT is expected to reach the desired target throughput during 780 the sustain phase. In addition, the measured KPIs must meet all 781 acceptance criteria. Follow the step 3, if the KPI metrics do not 782 meet the acceptance criteria. 784 7.1.4.3. Step 3: Test Iteration with Binary Search 786 Use binary search algorithm to configure the desired traffic load 787 profile for each test iteration. 789 Determine the maximum and average achievable throughput within the 790 acceptance criteria. 792 7.1.4.3.1. Pseudocode for binary search algorithm 794 TBD Resolution:=0.01* Target throughput and Backoff:= 50% 796 7.2. TCP Concurrent Connection Capacity 798 7.3. TCP Connection Setup Rate 800 7.4. Application Transaction Rate 802 7.5. SSL/TLS Handshake Rate 804 8. Formal Syntax 806 9. IANA Considerations 808 This document makes no request of IANA. 810 Note to RFC Editor: this section may be removed on publication as an 811 RFC. 813 10. Security Considerations 815 11. Acknowledgements 817 12. Normative References 819 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 820 Requirement Levels", BCP 14, RFC 2119, 821 DOI 10.17487/RFC2119, March 1997, 822 . 824 Appendix A. An Appendix 826 tbd 828 Author's Address 830 Balamuhunthan Balarajah 831 EANTC AG 832 Salzufer 14 833 Berlin 10587 834 Germany 836 Email: balarajah@eantc.de