1 CHARACTERIZE VSWITCH PERFORMANCE FOR TELCO NFV USE CASES LEVEL TEST DESIGN
2 ==========================================================================
4 .. contents:: Table of Contents
9 The objective of the OPNFV project titled
10 **“Characterize vSwitch Performance for Telco NFV Use Cases”**, is to
11 evaluate a virtual switch to identify its suitability for a Telco
12 Network Function Virtualization (NFV) environment. The intention of this
13 Level Test Design (LTD) document is to specify the set of tests to carry
14 out in order to objectively measure the current characteristics of a
15 virtual switch in the Network Function Virtualization Infrastructure
16 (NFVI) as well as the test pass criteria. The detailed test cases will
17 be defined in `Section 2 <#DetailsOfTheLevelTestDesign>`__, preceded by
18 the `Document identifier <#DocId>`__ and the `Scope <#Scope>`__.
20 This document is currently in draft form.
22 1.1. Document identifier
23 ------------------------
25 The document id will be used to uniquely
26 identify versions of the LTD. The format for the document id will be:
27 OPNFV\_vswitchperf\_LTD\_ver\_NUM\_MONTH\_YEAR\_STATUS, where by the
28 status is one of: draft, reviewed, corrected or final. The document id
29 for this version of the LTD is:
30 OPNFV\_vswitchperf\_LTD\_ver\_1.6\_Jan\_15\_DRAFT.
35 The main purpose of this project is to specify a suite of
36 performance tests in order to objectively measure the current packet
37 transfer characteristics of a virtual switch in the NFVI. The intent of
38 the project is to facilitate testing of any virtual switch. Thus, a
39 generic suite of tests shall be developed, with no hard dependencies to
40 a single implementation. In addition, the test case suite shall be
41 architecture independent.
43 The test cases developed in this project shall not form part of a
44 separate test framework, all of these tests may be inserted into the
45 Continuous Integration Test Framework and/or the Platform Functionality
46 Test Framework - if a vSwitch becomes a standard component of an OPNFV
52 * `RFC 1242 Benchmarking Terminology for Network Interconnection
53 Devices <http://www.ietf.org/rfc/rfc1242.txt>`__
54 * `RFC 2544 Benchmarking Methodology for Network Interconnect
55 Devices <http://www.ietf.org/rfc/rfc2544.txt>`__
56 * `RFC 2285 Benchmarking Terminology for LAN Switching
57 Devices <http://www.ietf.org/rfc/rfc2285.txt>`__
58 * `RFC 2889 Benchmarking Methodology for LAN Switching
59 Devices <http://www.ietf.org/rfc/rfc2889.txt>`__
60 * `RFC 3918 Methodology for IP Multicast
61 Benchmarking <http://www.ietf.org/rfc/rfc3918.txt>`__
62 * `RFC 4737 Packet Reordering
63 Metrics <http://www.ietf.org/rfc/rfc4737.txt>`__
64 * `RFC 5481 Packet Delay Variation Applicability
65 Statement <http://www.ietf.org/rfc/rfc5481.txt>`__
66 * `RFC 6201 Device Reset
67 Characterization <http://tools.ietf.org/html/rfc6201>`__
69 2. Details of the Level Test Design
70 ===================================
72 This section describes the features to be tested (`cf. 2.1
73 <#FeaturesToBeTested>`__), the test approach (`cf. 2.2 <#Approach>`__);
74 it also identifies the sets of test cases or scenarios (`cf. 2.3
75 <#TestIdentification>`__) along with the pass/fail criteria (`cf. 2.4
76 <#PassFail>`__) and the test deliverables (`cf. 2.5 <#TestDeliverables>`__).
78 2.1. Features to be tested
79 --------------------------
81 Characterizing virtual switches (i.e. Device Under Test (DUT) in this document)
82 includes measuring the following performance metrics:
84 - **Throughput** as defined by `RFC1242
85 <https://www.rfc-editor.org/rfc/rfc1242.txt>`__: The maximum rate at which
86 **none** of the offered frames are dropped by the DUT. The maximum frame
87 rate and bit rate that can be transmitted by the DUT without any error
88 should be recorded. Note there is an equivalent bit rate and a specific
89 layer at which the payloads contribute to the bits. Errors and
90 improperly formed frames or packets are dropped.
91 - **Packet delay** introduced by the DUT and its cumulative effect on
92 E2E networks. Frame delay can be measured equivalently.
93 - **Packet delay variation**: measured from the perspective of the
94 VNF/application. Packet delay variation is sometimes called "jitter".
95 However, we will avoid the term "jitter" as the term holds different
96 meaning to different groups of people. In this document we will
97 simply use the term packet delay variation. The preferred form for this
98 metric is the PDV form of delay variation defined in `RFC5481
99 <https://www.rfc-editor.org/rfc/rfc5481.txt>`__. The most relevant
100 measurement of PDV considers the delay variation of a single user flow,
101 as this will be relevant to the size of end-system buffers to compensate
102 for delay variation. The measurement system's ability to store the
103 delays of individual packets in the flow of interest is a key factor
104 that determines the specific measurement method. At the outset, it is
105 ideal to view the complete PDV distribution. Systems that can capture
106 and store packets and their delays have the freedom to calculate the
107 reference minimum delay and to determine various quantiles of the PDV
108 distribution accurately (in post-measurement processing routines).
109 Systems without storage must apply algorithms to calculate delay and
110 statistical measurements on the fly. For example, a system may store
111 temporary estimates of the mimimum delay and the set of (100) packets
112 with the longest delays during measurement (to calculate a high quantile,
113 and update these sets with new values periodically.
114 In some cases, a limited number of delay histogram bins will be
115 available, and the bin limits will need to be set using results from
116 repeated experiments. See section 8 of `RFC5481
117 <https://www.rfc-editor.org/rfc/rfc5481.txt>`__.
118 - **Packet loss** (within a configured waiting time at the receiver): All
119 packets sent to the DUT should be accounted for.
120 - **Burst behaviour**: measures the ability of the DUT to buffer packets.
121 - **Packet re-ordering**: measures the ability of the device under test to
122 maintain sending order throughout transfer to the destination.
123 - **Packet correctness**: packets or Frames must be well-formed, in that
124 they include all required fields, conform to length requirements, pass
125 integrity checks, etc.
126 - **Availability and capacity** of the DUT i.e. when the DUT is fully “up”
129 - Includes power consumption of the CPU (in various power states) and
131 - Includes CPU utilization.
132 - Includes the number of NIC interfaces supported.
133 - Includes headroom of VM workload processing cores (i.e. available
140 In order to determine the packet transfer characteristics of a virtual
141 switch, the tests will be broken down into the following categories:
143 2.2.1 Test Categories
144 ----------------------
145 - **Throughput Tests** to measure the maximum forwarding rate (in
146 frames per second or fps) and bit rate (in Mbps) for a constant load
147 (as defined by `RFC1242 <https://www.rfc-editor.org/rfc/rfc1242.txt>`__)
148 without traffic loss.
149 - **Packet and Frame Delay Tests** to measure average, min and max
150 packet and frame delay for constant loads.
151 - **Stream Performance Tests** (TCP, UDP) to measure bulk data transfer
152 performance, i.e. how fast systems can send and receive data through
154 - **Request/Response Performance** Tests (TCP, UDP) the measure the
155 transaction rate through the switch.
156 - **Packet Delay Tests** to understand latency distribution for
157 different packet sizes and over an extended test run to uncover
159 - **Scalability Tests** to understand how the virtual switch performs
160 as the number of flows, active ports, complexity of the forwarding
161 logic's configuration... it has to deal with increases.
162 - **Control Path and Datapath Coupling** Tests, to understand how
163 closely coupled the datapath and the control path are as well as the
164 effect of this coupling on the performance of the DUT.
165 - **CPU and Memory Consumption Tests** to understand the virtual
166 switch’s footprint on the system, this includes:
171 * Time To Establish Flows Tests.
173 - **Noisy Neighbour Tests**, to understand the effects of resource
174 sharing on the performance of a virtual switch.
176 **Note:** some of the tests above can be conducted simultaneously where
177 the combined results would be insightful, for example Packet/Frame Delay
180 2.2.2 Deployment Scenarios
181 --------------------------
182 The following represents possible deployments which can help to
183 determine the performance of both the virtual switch and the datapath
186 Physical port → vSwitch → physical port
187 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
188 .. code-block:: console
191 +--------------------------------------------------+ |
192 | +--------------------+ | |
195 | +--------------+ +--------------+ | |
196 | | phy port | vSwitch | phy port | | |
197 +---+--------------+------------+--------------+---+ _|
201 +--------------------------------------------------+
203 | traffic generator |
205 +--------------------------------------------------+
208 Physical port → vSwitch → VNF → vSwitch → physical port
209 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
210 .. code-block:: console
213 +---------------------------------------------------+ |
215 | +-------------------------------------------+ | |
216 | | Application | | |
217 | +-------------------------------------------+ | |
221 | +---------------+ +---------------+ | |
222 | | logical port 0| | logical port 1| | |
223 +---+---------------+-----------+---------------+---+ _|
227 +---+---------------+----------+---------------+---+ |
228 | | logical port 0| | logical port 1| | |
229 | +---------------+ +---------------+ | |
233 | +--------------+ +--------------+ | |
234 | | phy port | vSwitch | phy port | | |
235 +---+--------------+------------+--------------+---+ _|
239 +--------------------------------------------------+
241 | traffic generator |
243 +--------------------------------------------------+
246 Physical port → vSwitch → VNF → vSwitch → VNF → vSwitch → physical port
247 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
249 .. code-block:: console
252 +----------------------+ +----------------------+ |
253 | Guest 1 | | Guest 2 | |
254 | +---------------+ | | +---------------+ | |
255 | | Application | | | | Application | | |
256 | +---------------+ | | +---------------+ | |
258 | | v | | | v | | Guests
259 | +---------------+ | | +---------------+ | |
260 | | logical ports | | | | logical ports | | |
261 | | 0 1 | | | | 0 1 | | |
262 +---+---------------+--+ +---+---------------+--+ _|
266 +---+---------------+---------+---------------+--+ |
267 | | 0 1 | | 3 4 | | |
268 | | logical ports | | logical ports | | |
269 | +---------------+ +---------------+ | |
271 | | L-----------------+ v | |
272 | +--------------+ +--------------+ | |
273 | | phy ports | vSwitch | phy ports | | |
274 +---+--------------+----------+--------------+---+ _|
278 +--------------------------------------------------+
280 | traffic generator |
282 +--------------------------------------------------+
285 Physical port → vSwitch → VNF
286 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
288 .. code-block:: console
291 +---------------------------------------------------+ |
293 | +-------------------------------------------+ | |
294 | | Application | | |
295 | +-------------------------------------------+ | |
299 | +---------------+ | |
300 | | logical port 0| | |
301 +---+---------------+-------------------------------+ _|
305 +---+---------------+------------------------------+ |
306 | | logical port 0| | |
307 | +---------------+ | |
311 | +--------------+ | |
312 | | phy port | vSwitch | |
313 +---+--------------+------------ -------------- ---+ _|
317 +--------------------------------------------------+
319 | traffic generator |
321 +--------------------------------------------------+
323 VNF → vSwitch → physical port
324 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
326 .. code-block:: console
329 +---------------------------------------------------+ |
331 | +-------------------------------------------+ | |
332 | | Application | | |
333 | +-------------------------------------------+ | |
337 | +---------------+ | |
338 | | logical port | | |
339 +-------------------------------+---------------+---+ _|
343 +------------------------------+---------------+---+ |
344 | | logical port | | |
345 | +---------------+ | |
349 | +--------------+ | |
350 | vSwitch | phy port | | |
351 +-------------------------------+--------------+---+ _|
355 +--------------------------------------------------+
357 | traffic generator |
359 +--------------------------------------------------+
361 VNF → vSwitch → VNF → vSwitch
362 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
364 .. code-block:: console
367 +-------------------------+ +-------------------------+ |
368 | Guest 1 | | Guest 2 | |
369 | +-----------------+ | | +-----------------+ | |
370 | | Application | | | | Application | | |
371 | +-----------------+ | | +-----------------+ | |
375 | +---------------+ | | +---------------+ | |
376 | | logical port 0| | | | logical port 0| | |
377 +-----+---------------+---+ +---+---------------+-----+ _|
381 +----+---------------+------------+---------------+-----+ |
382 | | port 0 | | port 1 | | |
383 | +---------------+ +---------------+ | |
386 | +--------------------+ | |
389 +-------------------------------------------------------+ _|
391 HOST 1(Physical port → virtual switch → VNF → virtual switch →
392 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
393 Physical port) → HOST 2(Physical port → virtual switch → VNF →
394 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
395 virtual switch → Physical port)
396 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
398 .. code-block:: console
401 +----------------------+ +----------------------+ |
402 | Guest 1 | | Guest 2 | |
403 | +---------------+ | | +---------------+ | |
404 | | Application | | | | Application | | |
405 | +---------------+ | | +---------------+ | |
407 | | v | | | v | | Guests
408 | +---------------+ | | +---------------+ | |
409 | | logical ports | | | | logical ports | | |
410 | | 0 1 | | | | 0 1 | | |
411 +---+---------------+--+ +---+---------------+--+ _|
415 +---+---------------+--+ +---+---------------+--+ |
416 | | 0 1 | | | | 3 4 | | |
417 | | logical ports | | | | logical ports | | |
418 | +---------------+ | | +---------------+ | |
419 | ^ | | | ^ | | | Hosts
421 | +--------------+ | | +--------------+ | |
422 | | phy ports | | | | phy ports | | |
423 +---+--------------+---+ +---+--------------+---+ _|
425 | +-----------------+ |
427 +--------------------------------------------------+
429 | traffic generator |
431 +--------------------------------------------------+
435 **Note:** For tests where the traffic generator and/or measurement
436 receiver are implemented on VM and connected to the virtual switch
437 through vNIC, the issues of shared resources and interactions between
438 the measurement devices and the device under test must be considered.
440 **Note:** Some RFC 2889 tests require a full-mesh sending and receiving
441 pattern involving more than two ports. This possibility is illustrated in the
442 Physical port → vSwitch → VNF → vSwitch → VNF → vSwitch → physical port
443 diagram above (with 2 sending and 2 receiving ports, though all ports
444 could be used bi-directionally).
446 2.2.3 General Methodology:
447 --------------------------
448 To establish the baseline performance of the virtual switch, tests would
449 initially be run with a simple workload in the VNF (the recommended
450 simple workload VNF would be `DPDK <http://www.dpdk.org/>`__'s testpmd
451 application forwarding packets in a VM or vloop\_vnf a simple kernel
452 module that forwards traffic between two network interfaces inside the
453 virtualized environment while bypassing the networking stack).
454 Subsequently, the tests would also be executed with a real Telco
455 workload running in the VNF, which would exercise the virtual switch in
456 the context of higher level Telco NFV use cases, and prove that its
457 underlying characteristics and behaviour can be measured and validated.
458 Suitable real Telco workload VNFs are yet to be identified.
460 2.2.3.1 Default Test Parameters
461 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
463 The following list identifies the default parameters for suite of
466 - Reference application: Simple forwarding or Open Source VNF.
467 - Frame size (bytes): 64, 128, 256, 512, 1024, 1280, 1518, 2K, 4k OR
468 Packet size based on use-case (e.g. RTP 64B, 256B).
469 - Reordering check: Tests should confirm that packets within a flow are
471 - Duplex: Unidirectional / Bidirectional. Default: Full duplex with
472 traffic transmitting in both directions, as network traffic generally
473 does not flow in a single direction. By default the data rate of
474 transmitted traffic should be the same in both directions, please
475 note that asymmetric traffic (e.g. downlink-heavy) tests will be
476 mentioned explicitly for the relevant test cases.
477 - Number of Flows: Default for non scalability tests is a single flow.
478 For scalability tests the goal is to test with maximum supported
479 flows but where possible will test up to 10 Million flows. Start with
480 a single flow and scale up. By default flows should be added
481 sequentially, tests that add flows simultaneously will explicitly
482 call out their flow addition behaviour. Packets are generated across
483 the flows uniformly with no burstiness.
484 - Traffic Types: UDP, SCTP, RTP, GTP and UDP traffic.
485 - Deployment scenarios are:
486 - Physical → virtual switch → physical.
487 - Physical → virtual switch → VNF → virtual switch → physical.
488 - Physical → virtual switch → VNF → virtual switch → VNF → virtual
490 - Physical → virtual switch → VNF.
491 - VNF → virtual switch → Physical.
492 - VNF → virtual switch → VNF.
494 Tests MUST have these parameters unless otherwise stated. **Test cases
495 with non default parameters will be stated explicitly**.
497 **Note**: For throughput tests unless stated otherwise, test
498 configurations should ensure that traffic traverses the installed flows
499 through the switch, i.e. flows are installed and have an appropriate
500 time out that doesn't expire before packet transmission starts.
502 2.2.3.2 Flow Classification
503 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
505 Virtual switches classify packets into flows by processing and matching
506 particular header fields in the packet/frame and/or the input port where
507 the packets/frames arrived. The vSwitch then carries out an action on
508 the group of packets that match the classification parameters. Thus a
509 flow is considered to be a sequence of packets that have a shared set of
510 header field values or have arrived on the same port and have the same
511 action applied to them. Performance results can vary based on the
512 parameters the vSwitch uses to match for a flow. The recommended flow
513 classification parameters for L3 vSwitch performance tests are: the
514 input port, the source IP address, the destination IP address and the
515 Ethernet protocol type field. It is essential to increase the flow
516 time-out time on a vSwitch before conducting any performance tests that
517 do not measure the flow set-up time. Normally the first packet of a
518 particular flow will install the flow in the vSwitch which adds an
519 additional latency, subsequent packets of the same flow are not subject
520 to this latency if the flow is already installed on the vSwitch.
522 2.2.3.3 Test Priority
523 ~~~~~~~~~~~~~~~~~~~~~
525 Tests will be assigned a priority in order to determine which tests
526 should be implemented immediately and which tests implementations
529 Priority can be of following types: - Urgent: Must be implemented
530 immediately. - High: Must be implemented in the next release. - Medium:
531 May be implemented after the release. - Low: May or may not be
537 The SUT should be configured to its "default" state. The
538 SUT's configuration or set-up must not change between tests in any way
539 other than what is required to do the test. All supported protocols must
540 be configured and enabled for each test set up.
542 2.2.3.4.1 Port Configuration
543 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
545 The DUT should be configured with n ports where
546 n is a multiple of 2. Half of the ports on the DUT should be used as
547 ingress ports and the other half of the ports on the DUT should be used
548 as egress ports. Where a DUT has more than 2 ports, the ingress data
549 streams should be set-up so that they transmit packets to the egress
550 ports in sequence so that there is an even distribution of traffic
551 across ports. For example, if a DUT has 4 ports 0(ingress), 1(ingress),
552 2(egress) and 3(egress), the traffic stream directed at port 0 should
553 output a packet to port 2 followed by a packet to port 3. The traffic
554 stream directed at port 1 should also output a packet to port 2 followed
555 by a packet to port 3.
557 2.2.3.4.2 Frame Formats
558 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
560 Frame formats Layer 2 (data link layer) protocols
561 ++++++++++++++++++++++++++++++++++++++++++++++++++
564 .. code-block:: console
566 +---------------------+--------------------+-----------+
567 | Ethernet Header | Payload | Check Sum |
568 +---------------------+--------------------+-----------+
569 |_____________________|____________________|___________|
570 14 Bytes 46 - 1500 Bytes 4 Bytes
572 Layer 3 (network layer) protocols
573 ++++++++++++++++++++++++++++++++++
577 .. code-block:: console
579 +---------------------+--------------------+--------------------+-----------+
580 | Ethernet Header | IP Header | Payload | Check Sum |
581 +---------------------+--------------------+--------------------+-----------+
582 |_____________________|____________________|____________________|___________|
583 14 Bytes 20 bytes 26 - 1480 Bytes 4 Bytes
587 .. code-block:: console
589 +---------------------+--------------------+--------------------+-----------+
590 | Ethernet Header | IP Header | Payload | Check Sum |
591 +---------------------+--------------------+--------------------+-----------+
592 |_____________________|____________________|____________________|___________|
593 14 Bytes 40 bytes 26 - 1460 Bytes 4 Bytes
595 Layer 4 (transport layer) protocols
596 ++++++++++++++++++++++++++++++++++++
601 .. code-block:: console
603 +---------------------+--------------------+-----------------+--------------------+-----------+
604 | Ethernet Header | IP Header | Layer 4 Header | Payload | Check Sum |
605 +---------------------+--------------------+-----------------+--------------------+-----------+
606 |_____________________|____________________|_________________|____________________|___________|
607 14 Bytes 40 bytes 20 Bytes 6 - 1460 Bytes 4 Bytes
609 Layer 5 (application layer) protocols
610 +++++++++++++++++++++++++++++++++++++
614 .. code-block:: console
616 +---------------------+--------------------+-----------------+--------------------+-----------+
617 | Ethernet Header | IP Header | Layer 4 Header | Payload | Check Sum |
618 +---------------------+--------------------+-----------------+--------------------+-----------+
619 |_____________________|____________________|_________________|____________________|___________|
620 14 Bytes 20 bytes 20 Bytes Min 6 Bytes 4 Bytes
623 2.2.3.4.3 Packet Throughput
624 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
625 There is a difference between an Ethernet frame,
626 an IP packet, and a UDP datagram. In the seven-layer OSI model of
627 computer networking, packet refers to a data unit at layer 3 (network
628 layer). The correct term for a data unit at layer 2 (data link layer) is
629 a frame, and at layer 4 (transport layer) is a segment or datagram.
631 Important concepts related to 10GbE performance are frame rate and
632 throughput. The MAC bit rate of 10GbE, defined in the IEEE standard 802
633 .3ae, is 10 billion bits per second. Frame rate is based on the bit rate
634 and frame format definitions. Throughput, defined in IETF RFC 1242, is
635 the highest rate at which the system under test can forward the offered
638 The frame rate for 10GbE is determined by a formula that divides the 10
639 billion bits per second by the preamble + frame length + inter-frame
642 The maximum frame rate is calculated using the minimum values of the
643 following parameters, as described in the IEEE 802 .3ae standard:
645 - Preamble: 8 bytes \* 8 = 64 bits
646 - Frame Length: 64 bytes (minimum) \* 8 = 512 bits
647 - Inter-frame Gap: 12 bytes (minimum) \* 8 = 96 bits
649 Therefore, Maximum Frame Rate (64B Frames)
650 = MAC Transmit Bit Rate / (Preamble + Frame Length + Inter-frame Gap)
651 = 10,000,000,000 / (64 + 512 + 96)
652 = 10,000,000,000 / 672
653 = 14,880,952.38 frame per second (fps)
655 2.2.3.4.4 System isolation and validation
656 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
658 A key consideration when conducting any sort of benchmark is trying to
659 ensure the consistency and repeatability of test results between runs.
660 When benchmarking the performance of a virtual switch there are many
661 factors that can affect the consistency of results. This section
662 describes these factors and the measures that can be taken to limit
663 their effects. In addition, this section will outline some system tests
664 to validate the platform and the VNF before conducting any vSwitch
669 When conducting a benchmarking test on any SUT, it is essential to limit
670 (and if reasonable, eliminate) any noise that may interfere with the
671 accuracy of the metrics collected by the test. This noise may be
672 introduced by other hardware or software (OS, other applications), and
673 can result in significantly varying performance metrics being collected
674 between consecutive runs of the same test. In the case of characterizing
675 the performance of a virtual switch, there are a number of configuration
676 parameters that can help increase the repeatability and stability of
677 test results, including:
679 - OS/GRUB configuration:
681 - maxcpus = n where n >= 0; limits the kernel to using 'n'
682 processors. Only use exactly what you need.
683 - isolcpus: Isolate CPUs from the general scheduler. Isolate all
684 CPUs bar one which will be used by the OS.
685 - use taskset to affinitize the forwarding application and the VNFs
686 onto isolated cores. VNFs and the vSwitch should be allocated
687 their own cores, i.e. must not share the same cores. vCPUs for the
688 VNF should be affinitized to individual cores also.
689 - Limit the amount of background applications that are running and
690 set OS to boot to runlevel 3. Make sure to kill any unnecessary
691 system processes/daemons.
692 - Only enable hardware that you need to use for your test – to
693 ensure there are no other interrupts on the system.
694 - Configure NIC interrupts to only use the cores that are not
695 allocated to any other process (VNF/vSwitch).
697 - NUMA configuration: Any unused sockets in a multi-socket system
699 - CPU pinning: The vSwitch and the VNF should each be affinitized to
700 separate logical cores using a combination of maxcpus, isolcpus and
702 - BIOS configuration: BIOS should be configured for performance where
703 an explicit option exists, sleep states should be disabled, any
704 virtualization optimization technologies should be enabled, and
705 hyperthreading should also be enabled.
709 System validation is broken down into two sub-categories: Platform
710 validation and VNF validation. The validation test itself involves
711 verifying the forwarding capability and stability for the sub-system
712 under test. The rationale behind system validation is two fold. Firstly
713 to give a tester confidence in the stability of the platform or VNF that
714 is being tested; and secondly to provide base performance comparison
715 points to understand the overhead introduced by the virtual switch.
717 * Benchmark platform forwarding capability: This is an OPTIONAL test
718 used to verify the platform and measure the base performance (maximum
719 forwarding rate in fps and latency) that can be achieved by the
720 platform without a vSwitch or a VNF. The following diagram outlines
721 the set-up for benchmarking Platform forwarding capability:
723 .. code-block:: console
726 +--------------------------------------------------+ |
727 | +------------------------------------------+ | |
729 | | l2fw or DPDK L2FWD app | | Host
731 | +------------------------------------------+ | |
733 +---+------------------------------------------+---+ __|
737 +--------------------------------------------------+
739 | traffic generator |
741 +--------------------------------------------------+
743 * Benchmark VNF forwarding capability: This test is used to verify
744 the VNF and measure the base performance (maximum forwarding rate in
745 fps and latency) that can be achieved by the VNF without a vSwitch.
746 The performance metrics collected by this test will serve as a key
747 comparison point for NIC passthrough technologies and vSwitches. VNF
748 in this context refers to the hypervisor and the VM. The following
749 diagram outlines the set-up for benchmarking VNF forwarding
752 .. code-block:: console
755 +--------------------------------------------------+ |
756 | +------------------------------------------+ | |
760 | +------------------------------------------+ | |
761 | | Passthrough/SR-IOV | | Host
762 | +------------------------------------------+ | |
764 +---+------------------------------------------+---+ __|
768 +--------------------------------------------------+
770 | traffic generator |
772 +--------------------------------------------------+
775 Methodology to benchmark Platform/VNF forwarding capability
776 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
778 The recommended methodology for the platform/VNF validation and
779 benchmark is: - Run `RFC2889 <https://www.rfc-editor.org/rfc/rfc2289.txt>`__
780 Maximum Forwarding Rate test, this test will produce maximum
781 forwarding rate and latency results that will serve as the
782 expected values. These expected values can be used in
783 subsequent steps or compared with in subsequent validation tests. -
784 Transmit bidirectional traffic at line rate/max forwarding rate
785 (whichever is higher) for at least 72 hours, measure throughput (fps)
786 and latency. - Note: Traffic should be bidirectional. - Establish a
787 baseline forwarding rate for what the platform can achieve. - Additional
788 validation: After the test has completed for 72 hours run bidirectional
789 traffic at the maximum forwarding rate once more to see if the system is
790 still functional and measure throughput (fps) and latency. Compare the
791 measure the new obtained values with the expected values.
793 **NOTE 1**: How the Platform is configured for its forwarding capability
794 test (BIOS settings, GRUB configuration, runlevel...) is how the
795 platform should be configured for every test after this
797 **NOTE 2**: How the VNF is configured for its forwarding capability test
798 (# of vCPUs, vNICs, Memory, affinitization…) is how it should be
799 configured for every test that uses a VNF after this.
801 2.2.4 RFCs for testing switch performance
802 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
804 The starting point for defining the suite of tests for benchmarking the
805 performance of a virtual switch is to take existing RFCs and standards
806 that were designed to test their physical counterparts and adapting them
807 for testing virtual switches. The rationale behind this is to establish
808 a fair comparison between the performance of virtual and physical
809 switches. This section outlines the RFCs that are used by this
812 RFC 1242 Benchmarking Terminology for Network Interconnection
813 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
814 Devices RFC 1242 defines the terminology that is used in describing
815 performance benchmarking tests and their results. Definitions and
816 discussions covered include: Back-to-back, bridge, bridge/router,
817 constant load, data link frame size, frame loss rate, inter frame gap,
818 latency, and many more.
820 RFC 2544 Benchmarking Methodology for Network Interconnect Devices
821 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
822 RFC 2544 outlines a benchmarking methodology for network Interconnect
823 Devices. The methodology results in performance metrics such as latency,
824 frame loss percentage, and maximum data throughput.
826 In this document network “throughput” (measured in millions of frames
827 per second) is based on RFC 2544, unless otherwise noted. Frame size
828 refers to Ethernet frames ranging from smallest frames of 64 bytes to
829 largest frames of 4K bytes.
833 1. Throughput test defines the maximum number of frames per second
834 that can be transmitted without any error.
836 2. Latency test measures the time required for a frame to travel from
837 the originating device through the network to the destination device.
838 Please note that RFC2544 Latency measurement will be superseded with
839 a measurement of average latency over all successfully transferred
842 3. Frame loss test measures the network’s
843 response in overload conditions - a critical indicator of the
844 network’s ability to support real-time applications in which a
845 large amount of frame loss will rapidly degrade service quality.
847 4. Burst test assesses the buffering capability of a switch. It
848 measures the maximum number of frames received at full line rate
849 before a frame is lost. In carrier Ethernet networks, this
850 measurement validates the excess information rate (EIR) as defined in
853 5. System recovery to characterize speed of recovery from an overload
856 6. Reset to characterize speed of recovery from device or software
857 reset. This type of test has been updated by `RFC6201 <https://www.rfc-editor.org/rfc/rfc6201.txt>`__ as such,
858 the methodology defined by this specification will be that of RFC 6201.
860 Although not included in the defined RFC 2544 standard, another crucial
861 measurement in Ethernet networking is packet delay variation. The
862 definition set out by this specification comes from
863 `RFC5481 <https://www.rfc-editor.org/rfc/rfc5481.txt>`__.
865 RFC 2285 Benchmarking Terminology for LAN Switching Devices
866 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
867 RFC 2285 defines the terminology that is used to describe the
868 terminology for benchmarking a LAN switching device. It extends RFC
869 1242 and defines: DUTs, SUTs, Traffic orientation and distribution,
870 bursts, loads, forwarding rates, etc.
872 RFC 2889 Benchmarking Methodology for LAN Switching
873 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
874 RFC 2889 outlines a benchmarking methodology for LAN switching, it
875 extends RFC 2544. The outlined methodology gathers performance
876 metrics for forwarding, congestion control, latency, address handling
877 and finally filtering.
879 RFC 3918 Methodology for IP Multicast Benchmarking
880 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
881 RFC 3918 outlines a methodology for IP Multicast benchmarking.
883 RFC 4737 Packet Reordering Metrics
884 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
885 RFC 4737 describes metrics for identifying and counting re-ordered
886 packets within a stream, and metrics to measure the extent each
887 packet has been re-ordered.
889 RFC 5481 Packet Delay Variation Applicability Statement
890 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
891 RFC 5481 defined two common, but different forms of delay variation
892 metrics, and compares the metrics over a range of networking
893 circumstances and tasks. The most suitable form for vSwitch
894 benchmarking is the "PDV" form.
896 RFC 6201 Device Reset Characterization
897 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
898 RFC 6201 extends the methodology for characterizing the speed of
899 recovery of the DUT from device or software reset described in RFC
902 2.2.5 Details of the Test Report
903 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
905 There are a number of parameters related to the system, DUT and tests
906 that can affect the repeatability of a test results and should be
907 recorded. In order to minimise the variation in the results of a test,
908 it is recommended that the test report includes the following information:
910 - Hardware details including:
914 - Memory information (see below)
915 - Number of enabled cores.
916 - Number of cores used for the test.
917 - Number of physical NICs, as well as their details (manufacturer,
918 versions, type and the PCI slot they are plugged into).
919 - NIC interrupt configuration.
920 - BIOS version, release date and any configurations that were
923 - Software details including:
925 - OS version (for host and VNF)
926 - Kernel version (for host and VNF)
927 - GRUB boot parameters (for host and VNF).
928 - Hypervisor details (Type and version).
929 - Selected vSwitch, version number or commit id used.
930 - vSwitch launch command line if it has been parameterised.
931 - Memory allocation to the vSwitch – which NUMA node it is using,
932 and how many memory channels.
933 - Where the vswitch is built from source: compiler details including
934 versions and the flags that were used to compile the vSwitch.
935 - DPDK or any other SW dependency version number or commit id used.
936 - Memory allocation to a VM - if it's from Hugpages/elsewhere.
937 - VM storage type: snapshot/independent persistent/independent
940 - Number of Virtual NICs (vNICs), versions, type and driver.
941 - Number of virtual CPUs and their core affinity on the host.
942 - Number vNIC interrupt configuration.
943 - Thread affinitization for the applications (including the vSwitch
945 - Details of Resource isolation, such as CPUs designated for
946 Host/Kernel (isolcpu) and CPUs designated for specific processes
965 - Traffic Information:
967 - Traffic type - UDP, TCP, IMIX / Other.
970 - Deployment Scenario.
972 **Note**: Tests that require additional parameters to be recorded will
973 explicitly specify this.
975 2.3. Test identification
976 ------------------------
977 2.3.1 Throughput tests
978 ~~~~~~~~~~~~~~~~~~~~~~
979 The following tests aim to determine the maximum forwarding rate that
980 can be achieved with a virtual switch. The list is not exhaustive but
981 should indicate the type of tests that should be required. It is
982 expected that more will be added.
984 Test ID: LTD.Throughput.RFC2544.PacketLossRatio
985 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
986 **Title**: RFC 2544 X% packet loss ratio Throughput and Latency Test
988 **Prerequisite Test**: N/A
994 This test determines the DUT's maximum forwarding rate with X% traffic
995 loss for a constant load (fixed length frames at a fixed interval time).
996 The default loss percentages to be tested are: - X = 0% - X = 10^-7%
998 Note: Other values can be tested if required by the user.
1000 The selected frame sizes are those previously defined under `Default
1001 Test Parameters <#DefaultParams>`__. The test can also be used to
1002 determine the average latency of the traffic.
1004 Under the `RFC2544 <https://www.rfc-editor.org/rfc/rfc2544.txt>`__
1005 test methodology, the test duration will
1006 include a number of trials; each trial should run for a minimum period
1007 of 60 seconds. A binary search methodology must be applied for each
1008 trial to obtain the final result.
1010 **Expected Result**: At the end of each trial, the presence or absence
1011 of loss determines the modification of offered load for the next trial,
1012 converging on a maximum rate, or
1013 `RFC2544 <https://www.rfc-editor.org/rfc/rfc2544.txt>`__ Throughput with X% loss.
1014 The Throughput load is re-used in related
1015 `RFC2544 <https://www.rfc-editor.org/rfc/rfc2544.txt>`__ tests and other
1018 **Metrics Collected**:
1020 The following are the metrics collected for this test:
1022 - The maximum forwarding rate in Frames Per Second (FPS) and Mbps of
1023 the DUT for each frame size with X% packet loss.
1024 - The average latency of the traffic flow when passing through the DUT
1025 (if testing for latency, note that this average is different from the
1026 test specified in Section 26.3 of
1027 `RFC2544 <https://www.rfc-editor.org/rfc/rfc2544.txt>`__).
1028 - CPU and memory utilization may also be collected as part of this
1029 test, to determine the vSwitch's performance footprint on the system.
1031 Test ID: LTD.Throughput.RFC2544.PacketLossRatioFrameModification
1032 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1033 **Title**: RFC 2544 X% packet loss Throughput and Latency Test with
1036 **Prerequisite Test**: N/A
1042 This test determines the DUT's maximum forwarding rate with X% traffic
1043 loss for a constant load (fixed length frames at a fixed interval time).
1044 The default loss percentages to be tested are: - X = 0% - X = 10^-7%
1046 Note: Other values can be tested if required by the user.
1048 The selected frame sizes are those previously defined under `Default
1049 Test Parameters <#DefaultParams>`__. The test can also be used to
1050 determine the average latency of the traffic.
1052 Under the `RFC2544 <https://www.rfc-editor.org/rfc/rfc2544.txt>`__
1053 test methodology, the test duration will
1054 include a number of trials; each trial should run for a minimum period
1055 of 60 seconds. A binary search methodology must be applied for each
1056 trial to obtain the final result.
1058 During this test, the DUT must perform the following operations on the
1061 - Perform packet parsing on the DUT's ingress port.
1062 - Perform any relevant address look-ups on the DUT's ingress ports.
1063 - Modify the packet header before forwarding the packet to the DUT's
1064 egress port. Packet modifications include:
1066 - Modifying the Ethernet source or destination MAC address.
1067 - Modifying/adding a VLAN tag. (**Recommended**).
1068 - Modifying/adding a MPLS tag.
1069 - Modifying the source or destination ip address.
1070 - Modifying the TOS/DSCP field.
1071 - Modifying the source or destination ports for UDP/TCP/SCTP.
1072 - Modifying the TTL.
1074 **Expected Result**: The Packet parsing/modifications require some
1075 additional degree of processing resource, therefore the
1076 `RFC2544 <https://www.rfc-editor.org/rfc/rfc2544.txt>`__
1077 Throughput is expected to be somewhat lower than the Throughput level
1078 measured without additional steps. The reduction is expected to be
1079 greatest on tests with the smallest packet sizes (greatest header
1082 **Metrics Collected**:
1084 The following are the metrics collected for this test:
1086 - The maximum forwarding rate in Frames Per Second (FPS) and Mbps of
1087 the DUT for each frame size with X% packet loss and packet
1088 modification operations being performed by the DUT.
1089 - The average latency of the traffic flow when passing through the DUT
1090 (if testing for latency, note that this average is different from the
1091 test specified in Section 26.3 of
1092 `RFC2544 <https://www.rfc-editor.org/rfc/rfc2544.txt>`__).
1093 - The `RFC5481 <https://www.rfc-editor.org/rfc/rfc5481.txt>`__
1094 PDV form of delay variation on the traffic flow,
1095 using the 99th percentile.
1096 - CPU and memory utilization may also be collected as part of this
1097 test, to determine the vSwitch's performance footprint on the system.
1099 Test ID: LTD.Throughput.RFC2544.Profile
1100 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1101 **Title**: RFC 2544 Throughput and Latency Profile
1103 **Prerequisite Test**: N/A
1109 This test reveals how throughput and latency degrades as the offered
1110 rate varies in the region of the DUT's maximum forwarding rate as
1111 determined by LTD.Throughput.RFC2544.PacketLossRatio (0% Packet Loss).
1112 For example it can be used to determine if the degradation of throughput
1113 and latency as the offered rate increases is slow and graceful or sudden
1116 The selected frame sizes are those previously defined under `Default
1117 Test Parameters <#DefaultParams>`__.
1119 The offered traffic rate is described as a percentage delta with respect
1120 to the DUT's maximum forwarding rate as determined by
1121 LTD.Throughput.RFC2544.PacketLoss Ratio (0% Packet Loss case). A delta
1122 of 0% is equivalent to an offered traffic rate equal to the maximum
1123 forwarding rate; A delta of +50% indicates an offered rate half-way
1124 between the maximum forwarding rate and line-rate, whereas a delta of
1125 -50% indicates an offered rate of half the maximum rate. Therefore the
1126 range of the delta figure is natuarlly bounded at -100% (zero offered
1127 traffic) and +100% (traffic offered at line rate).
1129 The following deltas to the maximum forwarding rate should be applied:
1131 - -50%, -10%, 0%, +10% & +50%
1133 **Expected Result**: For each packet size a profile should be produced
1134 of how throughput and latency vary with offered rate.
1136 **Metrics Collected**:
1138 The following are the metrics collected for this test:
1140 - The forwarding rate in Frames Per Second (FPS) and Mbps of the DUT
1141 for each delta to the maximum forwarding rate and for each frame
1143 - The average latency for each delta to the maximum forwarding rate and
1144 for each frame size.
1145 - CPU and memory utilization may also be collected as part of this
1146 test, to determine the vSwitch's performance footprint on the system.
1147 - Any failures experienced (for example if the vSwitch crashes, stops
1148 processing packets, restarts or becomes unresponsive to commands)
1149 when the offered load is above Maximum Throughput MUST be recorded
1150 and reported with the results.
1152 Test ID: LTD.Throughput.RFC2544.SystemRecoveryTime
1153 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1154 **Title**: RFC 2544 System Recovery Time Test
1156 **Prerequisite Test** LTD.Throughput.RFC2544.PacketLossRatio
1162 The aim of this test is to determine the length of time it takes the DUT
1163 to recover from an overload condition for a constant load (fixed length
1164 frames at a fixed interval time). The selected frame sizes are those
1165 previously defined under `Default Test Parameters <#DefaultParams>`__,
1166 traffic should be sent to the DUT under normal conditions. During the
1167 duration of the test and while the traffic flows are passing though the
1168 DUT, at least one situation leading to an overload condition for the DUT
1169 should occur. The time from the end of the overload condition to when
1170 the DUT returns to normal operations should be measured to determine
1171 recovery time. Prior to overloading the DUT, one should record the
1172 average latency for 10,000 packets forwarded through the DUT.
1174 The overload condition SHOULD be to transmit traffic at a very high
1175 frame rate to the DUT (150% of the maximum 0% packet loss rate as
1176 determined by LTD.Throughput.RFC2544.PacketLossRatio or line-rate
1177 whichever is lower), for at least 60 seconds, then reduce the frame rate
1178 to 75% of the maximum 0% packet loss rate. A number of time-stamps
1179 should be recorded: - Record the time-stamp at which the frame rate was
1180 reduced and record a second time-stamp at the time of the last frame
1181 lost. The recovery time is the difference between the two timestamps. -
1182 Record the average latency for 10,000 frames after the last frame loss
1183 and continue to record average latency measurements for every 10,000
1184 frames, when latency returns to within 10% of pre-overload levels record
1187 **Expected Result**:
1189 **Metrics collected**
1191 The following are the metrics collected for this test:
1193 - The length of time it takes the DUT to recover from an overload
1195 - The length of time it takes the DUT to recover the average latency to
1196 pre-overload conditions.
1198 **Deployment scenario**:
1200 - Physical → virtual switch → physical.
1202 Test ID: LTD.Throughput.RFC2544.BackToBackFrames
1203 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1204 **Title**: RFC2544 Back To Back Frames Test
1206 **Prerequisite Test**: N
1212 The aim of this test is to characterize the ability of the DUT to
1213 process back-to-back frames. For each frame size previously defined
1214 under `Default Test Parameters <#DefaultParams>`__, a burst of traffic
1215 is sent to the DUT with the minimum inter-frame gap between each frame.
1216 If the number of received frames equals the number of frames that were
1217 transmitted, the burst size should be increased and traffic is sent to
1218 the DUT again. The value measured is the back-to-back value, that is the
1219 maximum burst size the DUT can handle without any frame loss.
1221 **Expected Result**:
1223 Tests of back-to-back frames with physical devices have produced
1224 unstable results in some cases. All tests should be repeated in multiple
1225 test sessions and results stability should be examined.
1227 **Metrics collected**
1229 The following are the metrics collected for this test:
1231 - The back-to-back value, which is the the number of frames in the
1232 longest burst that the DUT will handle without the loss of any
1234 - CPU and memory utilization may also be collected as part of this
1235 test, to determine the vSwitch's performance footprint on the system.
1237 **Deployment scenario**:
1239 - Physical → virtual switch → physical.
1241 Test ID: LTD.Throughput.RFC2889.Soak
1242 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1243 **Title**: RFC 2889 X% packet loss Throughput Soak Test
1245 **Prerequisite Test** LTD.Throughput.RFC2544.PacketLossRatio
1251 The aim of this test is to understand the Throughput stability over an
1252 extended test duration in order to uncover any outliers. To allow for an
1253 extended test duration, the test should ideally run for 24 hours or, if
1254 this is not possible, for at least 6 hours. For this test, each frame
1255 size must be sent at the highest Throughput with X% packet loss, as
1256 determined in the prerequisite test. The default loss percentages to be
1257 tested are: - X = 0% - X = 10^-7%
1259 Note: Other values can be tested if required by the user.
1261 **Expected Result**:
1263 **Metrics Collected**:
1265 The following are the metrics collected for this test:
1267 - Throughput stability of the DUT.
1269 - This means reporting the number of packets lost per time interval
1270 and reporting any time intervals with packet loss. The
1271 `RFC2889 <https://www.rfc-editor.org/rfc/rfc2289.txt>`__
1272 Forwarding Rate shall be measured in each interval.
1273 An interval of 60s is suggested.
1275 - CPU and memory utilization may also be collected as part of this
1276 test, to determine the vSwitch's performance footprint on the system.
1277 - The `RFC5481 <https://www.rfc-editor.org/rfc/rfc5481.txt>`__
1278 PDV form of delay variation on the traffic flow,
1279 using the 99th percentile.
1281 Test ID: LTD.Throughput.RFC2889.SoakFrameModification
1282 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1283 **Title**: RFC 2889 Throughput Soak Test with Frame Modification
1285 **Prerequisite Test**: LTD.Throughput.RFC2544.PacketLossRatioFrameModification (0% Packet Loss)
1291 The aim of this test is to understand the throughput stability over an
1292 extended test duration in order to uncover any outliers. To allow for an
1293 extended test duration, the test should ideally run for 24 hours or, if
1294 this is not possible, for at least 6 hour. For this test, each frame
1295 size must be sent at the highest Throughput with 0% packet loss, as
1296 determined in the prerequisite test.
1298 During this test, the DUT must perform the following operations on the
1301 - Perform packet parsing on the DUT's ingress port.
1302 - Perform any relevant address look-ups on the DUT's ingress ports.
1303 - Modify the packet header before forwarding the packet to the DUT's
1304 egress port. Packet modifications include:
1306 - Modifying the Ethernet source or destination MAC address.
1307 - Modifying/adding a VLAN tag (**Recommended**).
1308 - Modifying/adding a MPLS tag.
1309 - Modifying the source or destination ip address.
1310 - Modifying the TOS/DSCP field.
1311 - Modifying the source or destination ports for UDP/TCP/SCTP.
1312 - Modifying the TTL.
1314 **Expected Result**:
1316 **Metrics Collected**:
1318 The following are the metrics collected for this test:
1320 - Throughput stability of the DUT.
1322 - This means reporting the number of packets lost per time interval
1323 and reporting any time intervals with packet loss. The
1324 `RFC2889 <https://www.rfc-editor.org/rfc/rfc2289.txt>`__
1325 Forwarding Rate shall be measured in each interval.
1326 An interval of 60s is suggested.
1328 - CPU and memory utilization may also be collected as part of this
1329 test, to determine the vSwitch's performance footprint on the system.
1330 - The `RFC5481 <https://www.rfc-editor.org/rfc/rfc5481.txt>`__ PDV form of delay variation on the traffic flow,
1331 using the 99th percentile.
1333 Test ID: LTD.Throughput.RFC6201.ResetTime
1334 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1335 **Title**: RFC 6201 Reset Time Test
1337 **Prerequisite Test**: N/A
1343 The aim of this test is to determine the length of time it takes the DUT
1344 to recover from a reset.
1346 Two reset methods are defined - planned and unplanned. A planned reset
1347 requires stopping and restarting the virtual switch by the usual
1348 'graceful' method defined by it's documentation. An unplanned reset
1349 requires simulating a fatal internal fault in the virtual switch - for
1350 example by using kill -SIGKILL on a Linux environment.
1352 Both reset methods SHOULD be exercised.
1354 For each frame size previously defined under `Default Test
1355 Parameters <#DefaultParams>`__, traffic should be sent to the DUT under
1356 normal conditions. During the duration of the test and while the traffic
1357 flows are passing through the DUT, the DUT should be reset and the Reset
1358 time measured. The Reset time is the total time that a device is
1359 determined to be out of operation and includes the time to perform the
1360 reset and the time to recover from it (cf. `RFC6201 <https://www.rfc-editor.org/rfc/rfc6201.txt>`__).
1362 `RFC6201 <https://www.rfc-editor.org/rfc/rfc6201.txt>`__ defines two methods to measure the Reset time:
1363 - Frame-Loss Method: which requires the monitoring of the number of
1364 lost frames and calculates the Reset time based on the number of
1365 frames lost and the offered rate according to the following
1368 .. code-block:: console
1370 Frames_lost (packets)
1371 Reset_time = -------------------------------------
1372 Offered_rate (packets per second)
1374 - Timestamp Method: which measures the time from which the last frame
1375 is forwarded from the DUT to the time the first frame is forwarded
1376 after the reset. This involves time-stamping all transmitted frames
1377 and recording the timestamp of the last frame that was received prior
1378 to the reset and also measuring the timestamp of the first frame that
1379 is received after the reset. The Reset time is the difference between
1380 these two timestamps.
1382 According to `RFC6201 <https://www.rfc-editor.org/rfc/rfc6201.txt>`__ the choice of method depends on the test
1383 tool's capability; the Frame-Loss method SHOULD be used if the test tool
1384 supports: - Counting the number of lost frames per stream. -
1385 Transmitting test frame despite the physical link status.
1387 whereas the Timestamp method SHOULD be used if the test tool supports: -
1388 Timestamping each frame. - Monitoring received frame's timestamp. -
1389 Transmitting frames only if the physical link status is up.
1391 **Expected Result**:
1393 **Metrics collected**
1395 The following are the metrics collected for this test: - Average Reset
1396 Time over the number of trials performed.
1398 Results of this test should include the following information: - The
1399 reset method used. - Throughput in Fps and Mbps. - Average Frame Loss
1400 over the number of trials performed. - Average Reset Time in
1401 milliseconds over the number of trials performed. - Number of trials
1402 performed. - Protocol: IPv4, IPv6, MPLS, etc. - Frame Size in Octets -
1403 Port Media: Ethernet, Gigabit Ethernet (GbE), etc. - Port Speed: 10
1404 Gbps, 40 Gbps etc. - Interface Encapsulation: Ethernet, Ethernet VLAN,
1407 **Deployment scenario**:
1409 - Physical → virtual switch → physical.
1411 Test ID: LTD.Throughput.RFC2889.MaxForwardingRate
1412 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1413 **Title**: RFC2889 Forwarding Rate Test
1415 **Prerequisite Test**: LTD.Throughput.RFC2544.PacketLossRatio
1421 This test measures the DUT's Max Forwarding Rate when the Offered Load
1422 is varied between the throughput and the Maximum Offered Load for fixed
1423 length frames at a fixed time interval. The selected frame sizes are
1424 those previously defined under `Default Test
1425 Parameters <#DefaultParams>`__. The throughput is the maximum offered
1426 load with 0% frame loss (measured by the prerequisite test), and the
1427 Maximum Offered Load (as defined by
1428 `RFC2285 <https://www.rfc-editor.org/rfc/rfc2285.txt>`__) is *"the highest
1429 number of frames per second that an external source can transmit to a
1430 DUT/SUT for forwarding to a specified output interface or interfaces"*.
1432 Traffic should be sent to the DUT at a particular rate (TX rate)
1433 starting with TX rate equal to the throughput rate. The rate of
1434 successfully received frames at the destination counted (in FPS). If the
1435 RX rate is equal to the TX rate, the TX rate should be increased by a
1436 fixed step size and the RX rate measured again until the Max Forwarding
1439 The trial duration for each iteration should last for the period of time
1440 needed for the system to reach steady state for the frame size being
1441 tested. Under `RFC2889 <https://www.rfc-editor.org/rfc/rfc2289.txt>`__
1442 (Sec. 5.6.3.1) test methodology, the test
1443 duration should run for a minimum period of 30 seconds, regardless
1444 whether the system reaches steady state before the minimum duration
1447 **Expected Result**: According to
1448 `RFC2889 <https://www.rfc-editor.org/rfc/rfc2289.txt>`__ The Max Forwarding Rate
1449 is the highest forwarding rate of a DUT taken from an iterative set of
1450 forwarding rate measurements. The iterative set of forwarding rate
1451 measurements are made by setting the intended load transmitted from an
1452 external source and measuring the offered load (i.e what the DUT is
1453 capable of forwarding). If the Throughput == the Maximum Offered Load,
1454 it follows that Max Forwarding Rate is equal to the Maximum Offered
1457 **Metrics Collected**:
1459 The following are the metrics collected for this test:
1461 - The Max Forwarding Rate for the DUT for each packet size.
1462 - CPU and memory utilization may also be collected as part of this
1463 test, to determine the vSwitch's performance footprint on the system.
1465 Test ID: LTD.Throughput.RFC2889.ForwardPressure
1466 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1467 **Title**: RFC2889 Forward Pressure Test
1469 **Prerequisite Test**: LTD.Throughput.RFC2889.MaxForwardingRate
1475 The aim of this test is to determine if the DUT transmits frames with an
1476 inter-frame gap that is less than 12 bytes. This test overloads the DUT
1477 and measures the output for forward pressure. Traffic should be
1478 transmitted to the DUT with an inter-frame gap of 11 bytes, this will
1479 overload the DUT by 1 byte per frame. The forwarding rate of the DUT
1482 **Expected Result**: The forwarding rate should not exceed the maximum
1483 forwarding rate of the DUT collected by
1484 LTD.Throughput.RFC2889.MaxForwardingRate.
1486 **Metrics collected**
1488 The following are the metrics collected for this test:
1490 - Forwarding rate of the DUT in FPS or Mbps.
1491 - CPU and memory utilization may also be collected as part of this
1492 test, to determine the vSwitch's performance footprint on the system.
1494 **Deployment scenario**:
1496 - Physical → virtual switch → physical.
1498 Test ID: LTD.Throughput.RFC2889.AddressCachingCapacity
1499 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1500 **Title**: RFC2889 Address Caching Capacity Test
1502 **Prerequisite Test**: N/A
1508 Please note this test is only applicable to switches that are capable of
1509 MAC learning. The aim of this test is to determine the address caching
1510 capacity of the DUT for a constant load (fixed length frames at a fixed
1511 interval time). The selected frame sizes are those previously defined
1512 under `Default Test Parameters <#DefaultParams>`__.
1514 In order to run this test the aging time, that is the maximum time the
1515 DUT will keep a learned address in its flow table, and a set of initial
1516 addresses, whose value should be >= 1 and <= the max number supported by
1517 the implementation must be known. Please note that if the aging time is
1518 configurable it must be longer than the time necessary to produce frames
1519 from the external source at the specified rate. If the aging time is
1520 fixed the frame rate must be brought down to a value that the external
1521 source can produce in a time that is less than the aging time.
1523 Learning Frames should be sent from an external source to the DUT to
1524 install a number of flows. The Learning Frames must have a fixed
1525 destination address and must vary the source address of the frames. The
1526 DUT should install flows in its flow table based on the varying source
1527 addresses. Frames should then be transmitted from an external source at
1528 a suitable frame rate to see if the DUT has properly learned all of the
1529 addresses. If there is no frame loss and no flooding, the number of
1530 addresses sent to the DUT should be increased and the test is repeated
1531 until the max number of cached addresses supported by the DUT
1534 **Expected Result**:
1536 **Metrics collected**:
1538 The following are the metrics collected for this test:
1540 - Number of cached addresses supported by the DUT.
1541 - CPU and memory utilization may also be collected as part of this
1542 test, to determine the vSwitch's performance footprint on the system.
1544 **Deployment scenario**:
1546 - Physical → virtual switch → physical.
1548 Test ID: LTD.Throughput.RFC2889.AddressLearningRate
1549 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1550 **Title**: RFC2889 Address Learning Rate Test
1552 **Prerequisite Test**: LTD.Memory.RFC2889.AddressCachingCapacity
1558 Please note this test is only applicable to switches that are capable of
1559 MAC learning. The aim of this test is to determine the rate of address
1560 learning of the DUT for a constant load (fixed length frames at a fixed
1561 interval time). The selected frame sizes are those previously defined
1562 under `Default Test Parameters <#DefaultParams>`__, traffic should be
1563 sent with each IPv4/IPv6 address incremented by one. The rate at which
1564 the DUT learns a new address should be measured. The maximum caching
1565 capacity from LTD.Memory.RFC2889.AddressCachingCapacity should be taken
1566 into consideration as the maximum number of addresses for which the
1567 learning rate can be obtained.
1569 **Expected Result**: It may be worthwhile to report the behaviour when
1570 operating beyond address capacity - some DUTS may be more friendly to
1571 new addresses than others.
1573 **Metrics collected**:
1575 The following are the metrics collected for this test:
1577 - The address learning rate of the DUT.
1579 **Deployment scenario**:
1581 - Physical → virtual switch → physical.
1583 Test ID: LTD.Throughput.RFC2889.ErrorFramesFiltering
1584 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1585 **Title**: RFC2889 Error Frames Filtering Test
1587 **Prerequisite Test**: N/A
1593 The aim of this test is to determine whether the DUT will propagate any
1594 erroneous frames it receives or whether it is capable of filtering out
1595 the erroneous frames. Traffic should be sent with erroneous frames
1596 included within the flow at random intervals. Illegal frames that must
1597 be tested include: - Oversize Frames. - Undersize Frames. - CRC Errored
1598 Frames. - Dribble Bit Errored Frames - Alignment Errored Frames
1600 The traffic flow exiting the DUT should be recorded and checked to
1601 determine if the erroneous frames where passed through the DUT.
1603 **Expected Result**: Broken frames are not passed!
1605 **Metrics collected**
1607 No Metrics are collected in this test, instead it determines:
1609 - Whether the DUT will propagate erroneous frames.
1610 - Or whether the DUT will correctly filter out any erroneous frames
1611 from traffic flow with out removing correct frames.
1613 **Deployment scenario**:
1615 - Physical → virtual switch → physical.
1617 Test ID: LTD.Throughput.RFC2889.BroadcastFrameForwarding
1618 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1619 **Title**: RFC2889 Broadcast Frame Forwarding Test
1621 **Prerequisite Test**: N
1627 The aim of this test is to determine the maximum forwarding rate of the
1628 DUT when forwarding broadcast traffic. For each frame previously defined
1629 under `Default Test Parameters <#DefaultParams>`__, the traffic should
1630 be set up as broadcast traffic. The traffic throughput of the DUT should
1633 The test should be conducted with at least 4 physical ports on the DUT.
1634 The number of ports used MUST be recorded.
1636 As broadcast involves forwarding a single incoming packet to several
1637 destinations, the latency of a single packet is defined as the average
1638 of the latencies for each of the broadcast destinations.
1640 The incoming packet is transmitted on each of the other physical ports,
1641 it is not transmitted on the port on which it was received. The test MAY
1642 be conducted using different broadcasting ports to uncover any
1643 performance differences.
1645 **Expected Result**:
1647 **Metrics collected**:
1649 The following are the metrics collected for this test:
1651 - The forwarding rate of the DUT when forwarding broadcast traffic.
1652 - The minimum, average & maximum packets latencies observed.
1654 **Deployment scenario**:
1656 - Physical → virtual switch 3x physical.
1658 2.3.2 Packet Latency tests
1659 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
1660 These tests will measure the store and forward latency as well as the packet
1661 delay variation for various packet types through the virtual switch. The
1662 following list is not exhaustive but should indicate the type of tests
1663 that should be required. It is expected that more will be added.
1665 Test ID: LTD.PacketLatency.InitialPacketProcessingLatency
1666 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1667 **Title**: Initial Packet Processing Latency
1669 **Prerequisite Test**: N/A
1675 In some virtual switch architectures, the first packets of a flow will
1676 take the system longer to process than subsequent packets in the flow.
1677 This test determines the latency for these packets. The test will
1678 measure the latency of the packets as they are processed by the
1679 flow-setup-path of the DUT. There are two methods for this test, a
1680 recommended method and a nalternative method that can be used if it is
1681 possible to disable the fastpath of the virtual switch.
1683 Recommended method: This test will send 64,000 packets to the DUT, each
1684 belonging to a different flow. Average packet latency will be determined
1685 over the 64,000 packets.
1687 Alternative method: This test will send a single packet to the DUT after
1688 a fixed interval of time. The time interval will be equivalent to the
1689 amount of time it takes for a flow to time out in the virtual switch
1690 plus 10%. Average packet latency will be determined over 1,000,000
1693 This test is intended only for non-learning switches; For learning
1694 switches use RFC2889.
1696 For this test, only unidirectional traffic is required.
1698 **Expected Result**: The average latency for the initial packet of all
1699 flows should be greater than the latency of subsequent traffic.
1701 **Metrics Collected**:
1703 The following are the metrics collected for this test:
1705 - Average latency of the initial packets of all flows that are
1706 processed by the DUT.
1708 **Deployment scenario**:
1710 - Physical → Virtual Switch → Physical.
1712 Test ID: LTD.PacketDelayVariation.RFC3393.Soak
1713 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1714 **Title**: Packet Delay Variation Soak Test
1716 **Prerequisite Tests**: LTD.Throughput.RFC2544.PacketLossRatio (0% Packet Loss)
1722 The aim of this test is to understand the distribution of packet delay
1723 variation for different frame sizes over an extended test duration and
1724 to determine if there are any outliers. To allow for an extended test
1725 duration, the test should ideally run for 24 hours or, if this is not
1726 possible, for at least 6 hour. For this test, each frame size must be
1727 sent at the highest possible throughput with 0% packet loss, as
1728 determined in the prerequisite test.
1730 **Expected Result**:
1732 **Metrics Collected**:
1734 The following are the metrics collected for this test:
1736 - The packet delay variation value for traffic passing through the DUT.
1737 - The `RFC5481 <https://www.rfc-editor.org/rfc/rfc5481.txt>`__
1738 PDV form of delay variation on the traffic flow,
1739 using the 99th percentile, for each 60s interval during the test.
1740 - CPU and memory utilization may also be collected as part of this
1741 test, to determine the vSwitch's performance footprint on the system.
1743 2.3.3 Scalability tests
1744 ~~~~~~~~~~~~~~~~~~~~~~~~
1745 The general aim of these tests is to understand the impact of large flow
1746 table size and flow lookups on throughput. The following list is not
1747 exhaustive but should indicate the type of tests that should be required.
1748 It is expected that more will be added.
1750 Test ID: LTD.Scalability.RFC2544.0PacketLoss
1751 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1752 **Title**: RFC 2544 0% loss Scalability throughput test
1754 **Prerequisite Test**:
1760 The aim of this test is to measure how throughput changes as the number
1761 of flows in the DUT increases. The test will measure the throughput
1762 through the fastpath, as such the flows need to be installed on the DUT
1763 before passing traffic.
1765 For each frame size previously defined under `Default Test
1766 Parameters <#DefaultParams>`__ and for each of the following number of
1776 - Max supported number of flows.
1778 The maximum 0% packet loss throughput should be determined in a manner
1779 identical to LTD.Throughput.RFC2544.PacketLossRatio.
1781 **Expected Result**:
1783 **Metrics Collected**:
1785 The following are the metrics collected for this test:
1787 - The maximum number of frames per second that can be forwarded at the
1788 specified number of flows and the specified frame size, with zero
1791 Test ID: LTD.MemoryBandwidth.RFC2544.0PacketLoss.Scalability
1792 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1793 **Title**: RFC 2544 0% loss Memory Bandwidth Scalability test
1795 **Prerequisite Tests**:
1801 The aim of this test is to understand how the DUT's performance is
1802 affected by cache sharing and memory bandwidth between processes.
1804 During the test all cores not used by the vSwitch should be running a
1805 memory intensive application. This application should read and write
1806 random data to random addresses in unused physical memory. The random
1807 nature of the data and addresses is intended to consume cache, exercise
1808 main memory access (as opposed to cache) and exercise all memory buses
1809 equally. Furthermore: - the ratio of reads to writes should be recorded.
1810 A ratio of 1:1 SHOULD be used. - the reads and writes MUST be of
1811 cache-line size and be cache-line aligned. - in NUMA architectures
1812 memory access SHOULD be local to the core's node. Whether only local
1813 memory or a mix of local and remote memory is used MUST be recorded. -
1814 the memory bandwidth (reads plus writes) used per-core MUST be recorded;
1815 the test MUST be run with a per-core memory bandwidth equal to half the
1816 maximum system memory bandwidth divided by the number of cores. The test
1817 MAY be run with other values for the per-core memory bandwidth. - the
1818 test MAY also be run with the memory intensive application running on
1821 Under these conditions the DUT's 0% packet loss throughput is determined
1822 as per LTD.Throughput.RFC2544.PacketLossRatio.
1824 **Expected Result**:
1826 **Metrics Collected**:
1828 The following are the metrics collected for this test:
1830 - The DUT's 0% packet loss throughput in the presence of cache sharing and memory bandwidth between processes.
1832 2.3.5 Coupling between control path and datapath Tests
1833 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1834 The following tests aim to determine how tightly coupled the datapath
1835 and the control path are within a virtual switch. The following list
1836 is not exhaustive but should indicate the type of tests that should be
1837 required. It is expected that more will be added.
1839 Test ID: LTD.CPDPCouplingFlowAddition
1840 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1841 **Title**: Control Path and Datapath Coupling
1843 **Prerequisite Test**:
1849 The aim of this test is to understand how exercising the DUT's control
1850 path affects datapath performance.
1852 Initially a certain number of flow table entries are installed in the
1853 vSwitch. Then over the duration of an RFC2544 throughput test
1854 flow-entries are added and removed at the rates specified below. No
1855 traffic is 'hitting' these flow-entries, they are simply added and
1858 The test MUST be repeated with the following initial number of
1859 flow-entries installed: - < 10 - 1000 - 100,000 - 10,000,000 (or the
1860 maximum supported number of flow-entries)
1862 The test MUST be repeated with the following rates of flow-entry
1863 addition and deletion per second: - 0 - 1 (i.e. 1 addition plus 1
1864 deletion) - 100 - 10,000
1866 **Expected Result**:
1868 **Metrics Collected**:
1870 The following are the metrics collected for this test:
1872 - The maximum forwarding rate in Frames Per Second (FPS) and Mbps of
1874 - The average latency of the traffic flow when passing through the DUT
1875 (if testing for latency, note that this average is different from the
1876 test specified in Section 26.3 of
1877 `RFC2544 <https://www.rfc-editor.org/rfc/rfc2544.txt>`__).
1878 - CPU and memory utilization may also be collected as part of this
1879 test, to determine the vSwitch's performance footprint on the system.
1881 **Deployment scenario**:
1883 - Physical → virtual switch → physical.
1885 2.3.4 CPU and memory consumption
1886 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1887 The following tests will profile a virtual switch's CPU and memory
1888 utilization under various loads and circumstances. The following
1889 list is not exhaustive but should indicate the type of tests that
1890 should be required. It is expected that more will be added.
1892 Test ID: LTD.CPU.RFC2544.0PacketLoss
1893 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1894 **Title**: RFC 2544 0% Loss Compute Test
1896 **Prerequisite Test**:
1902 The aim of this test is to understand the overall performance of the
1903 system when a CPU intensive application is run on the same DUT as the
1904 Virtual Switch. For each frame size, an
1905 LTD.Throughput.RFC2544.PacketLossRatio (0% Packet Loss) test should be
1906 performed. Throughout the entire test a CPU intensive application should
1907 be run on all cores on the system not in use by the Virtual Switch. For
1908 NUMA system only cores on the same NUMA node are loaded.
1910 It is recommended that stress-ng be used for loading the non-Virtual
1911 Switch cores but any stress tool MAY be used.
1913 **Expected Result**:
1915 **Metrics Collected**:
1917 The following are the metrics collected for this test:
1919 - CPU utilization of the cores running the Virtual Switch.
1920 - The number of identity of the cores allocated to the Virtual Switch.
1921 - The configuration of the stress tool (for example the command line
1922 parameters used to start it.)
1924 2.3.9 Summary List of Tests
1925 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
1928 - Test ID: LTD.Throughput.RFC2544.PacketLossRatio
1929 - Test ID: LTD.Throughput.RFC2544.PacketLossRatioFrameModification
1930 - Test ID: LTD.Throughput.RFC2544.Profile
1931 - Test ID: LTD.Throughput.RFC2544.SystemRecoveryTime
1932 - Test ID: LTD.Throughput.RFC2544.BackToBackFrames
1933 - Test ID: LTD.Throughput.RFC2889.Soak
1934 - Test ID: LTD.Throughput.RFC2889.SoakFrameModification
1935 - Test ID: LTD.Throughput.RFC6201.ResetTime
1936 - Test ID: LTD.Throughput.RFC2889.MaxForwardingRate
1937 - Test ID: LTD.Throughput.RFC2889.ForwardPressure
1938 - Test ID: LTD.Throughput.RFC2889.AddressCachingCapacity
1939 - Test ID: LTD.Throughput.RFC2889.AddressLearningRate
1940 - Test ID: LTD.Throughput.RFC2889.ErrorFramesFiltering
1941 - Test ID: LTD.Throughput.RFC2889.BroadcastFrameForwarding
1943 2. Packet Latency tests
1945 - Test ID: LTD.PacketLatency.InitialPacketProcessingLatency
1946 - Test ID: LTD.PacketDelayVariation.RFC3393.Soak
1948 3. Scalability tests
1950 - Test ID: LTD.Scalability.RFC2544.0PacketLoss
1951 - Test ID: LTD.MemoryBandwidth.RFC2544.0PacketLoss.Scalability
1953 4. Coupling between control path and datapath Tests
1955 - Test ID: LTD.CPDPCouplingFlowAddition
1957 5. CPU and memory consumption
1959 - Test ID: LTD.CPU.RFC2544.0PacketLoss