diff options
Diffstat (limited to 'docs/report/testpmd_performance_tests_hw')
| -rw-r--r-- | docs/report/testpmd_performance_tests_hw/csit_release_notes.rst | 72 | ||||
| -rw-r--r-- | docs/report/testpmd_performance_tests_hw/overview.rst | 58 |
2 files changed, 44 insertions, 86 deletions
diff --git a/docs/report/testpmd_performance_tests_hw/csit_release_notes.rst b/docs/report/testpmd_performance_tests_hw/csit_release_notes.rst index 3b138a6f0c..cc1bf8dd1c 100644 --- a/docs/report/testpmd_performance_tests_hw/csit_release_notes.rst +++ b/docs/report/testpmd_performance_tests_hw/csit_release_notes.rst @@ -4,68 +4,10 @@ CSIT Release Notes Changes in CSIT |release| ------------------------- -#. Naming change for all Testpmd performance test suites and test cases. - #. Added Testpmd tests - new NICs - Intel x520 - -Performance Tests Naming ------------------------- - -CSIT |release| introduced a common structured naming convention for all -performance and functional tests. This change was driven by substantially -growing number and type of CSIT test cases. Firstly, the original practice did -not always follow any strict naming convention. Secondly test names did not -always clearly capture tested packet encapsulations, and the actual type or -content of the tests. Thirdly HW configurations in terms of NICs, ports and -their locality were not captured either. These were but few reasons that drove -the decision to change and define a new more complete and stricter test naming -convention, and to apply this to all existing and new test cases. - -The new naming should be intuitive for majority of the tests. The complete -description of CSIT test naming convention is provided on `CSIT test naming wiki -<https://wiki.fd.io/view/CSIT/csit-test-naming>`_. - -Here few illustrative examples of the new naming usage for performance test -suites: - -#. **Physical port to physical port - a.k.a. NIC-to-NIC, Phy-to-Phy, P2P** - - - *PortNICConfig-WireEncapsulation-PacketForwardingFunction- - PacketProcessingFunction1-...-PacketProcessingFunctionN-TestType* - - *10ge2p1x520-dot1q-l2bdbasemaclrn-ndrdisc.robot* => 2 ports of 10GE on - Intel x520 NIC, dot1q tagged Ethernet, L2 bridge-domain baseline switching - with MAC learning, NDR throughput discovery. - - *10ge2p1x520-ethip4vxlan-l2bdbasemaclrn-ndrchk.robot* => 2 ports of 10GE - on Intel x520 NIC, IPv4 VXLAN Ethernet, L2 bridge-domain baseline - switching with MAC learning, NDR throughput discovery. - - *10ge2p1x520-ethip4-ip4base-ndrdisc.robot* => 2 ports of 10GE on Intel - x520 NIC, IPv4 baseline routed forwarding, NDR throughput discovery. - - *10ge2p1x520-ethip6-ip6scale200k-ndrdisc.robot* => 2 ports of 10GE on - Intel x520 NIC, IPv6 scaled up routed forwarding, NDR throughput - discovery. - -#. **Physical port to VM (or VM chain) to physical port - a.k.a. NIC2VM2NIC, - P2V2P, NIC2VMchain2NIC, P2V2V2P** - - - *PortNICConfig-WireEncapsulation-PacketForwardingFunction- - PacketProcessingFunction1-...-PacketProcessingFunctionN-VirtEncapsulation- - VirtPortConfig-VMconfig-TestType* - - *10ge2p1x520-dot1q-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc.robot* => 2 ports - of 10GE on Intel x520 NIC, dot1q tagged Ethernet, L2 bridge-domain - switching to/from two vhost interfaces and one VM, NDR throughput - discovery. - - *10ge2p1x520-ethip4vxlan-l2bdbasemaclrn-eth-2vhost-1vm-ndrdisc.robot* => 2 - ports of 10GE on Intel x520 NIC, IPv4 VXLAN Ethernet, L2 bridge-domain - switching to/from two vhost interfaces and one VM, NDR throughput - discovery. - - *10ge2p1x520-ethip4vxlan-l2bdbasemaclrn-eth-4vhost-2vm-ndrdisc.robot* => 2 - ports of 10GE on Intel x520 NIC, IPv4 VXLAN Ethernet, L2 bridge-domain - switching to/from four vhost interfaces and two VMs, NDR throughput - discovery. - Multi-Thread and Multi-Core Measurements ---------------------------------------- @@ -79,12 +21,12 @@ HyperThreading Enabled (requires BIOS settings change and hard reboot). **Multi-core Test** - CSIT |release| multi-core tests are executed in the following Testpmd thread and core configurations: -#. 1t1c - 1 Testpmd worker thread on 1 CPU physical core. -#. 2t2c - 2 Testpmd worker threads on 2 CPU physical cores. -#. 4t4c - 4 Testpmd threads on 4 CPU physical cores. +#. 1t1c - 1 Testpmd pmd thread on 1 CPU physical core. +#. 2t2c - 2 Testpmd pmd threads on 2 CPU physical cores. +#. 4t4c - 4 Testpmd pmd threads on 4 CPU physical cores. -Note that in quite a few test cases running Testpmd on 2 or 4 physical cores -hits the tested NIC I/O bandwidth or packets-per-second limit. +Note that in many tests running Testpmd reaches tested NIC I/O bandwidth +or packets-per-second limit. Packet Throughput Measurements ------------------------------ @@ -137,8 +79,8 @@ Reported latency values are measured using following methodology: interface line. -Report Addendum Tests - More NICs ---------------------------------- +Report Addendum Tests - Additional NICs +--------------------------------------- Adding test cases with more NIC types. Once the results become available, they will be published as an addendum to the current version of CSIT |release| diff --git a/docs/report/testpmd_performance_tests_hw/overview.rst b/docs/report/testpmd_performance_tests_hw/overview.rst index 5bd81abac2..4ef3982e49 100644 --- a/docs/report/testpmd_performance_tests_hw/overview.rst +++ b/docs/report/testpmd_performance_tests_hw/overview.rst @@ -1,11 +1,13 @@ Overview ======== -Testpmd Performance Test Topologies ------------------------------------ +Tested Topologies HW +-------------------- CSIT Testpmd performance tests are executed on physical baremetal servers hosted -by LF FD.io project. Testbed physical topology is shown in the figure below.:: +by LF FD.io project. Testbed physical topology is shown in the figure below. + +:: +------------------------+ +------------------------+ | | | | @@ -25,13 +27,14 @@ by LF FD.io project. Testbed physical topology is shown in the figure below.:: | | +-----------+ -SUT1 and SUT2 are two System Under Test servers (Cisco UCS C240, each with two -Intel XEON CPUs), TG is a Traffic Generator (TG, another Cisco UCS C240, with -two Intel XEON CPUs). SUTs run Testpmd SW application in Linux user-mode as a -Device Under Test (DUT). TG runs TRex SW application as a packet Traffic -Generator. Physical connectivity between SUTs and to TG is provided using -different NIC models that need to be tested for performance. Currently -installed and tested NIC models include: +SUT1 and SUT2 are two System Under Test servers (currently Cisco UCS C240, +each with two Intel XEON CPUs), TG is a Traffic Generator (TG, currently +another Cisco UCS C240, with two Intel XEON CPUs). SUTs run Testpmd SW +application in Linux user-mode as a Device Under Test (DUT). TG runs TRex SW +application as a packet Traffic Generator. Physical connectivity between SUTs +and to TG is provided using direct links (no L2 switches) connecting different +NIC models that need to be tested for performance. Currently installed and +tested NIC models include: #. 2port10GE X520-DA2 Intel. #. 2port10GE X710 Intel. @@ -39,26 +42,39 @@ installed and tested NIC models include: #. 2port40GE VIC1385 Cisco. #. 2port40GE XL710 Intel. -Detailed LF FD.io test bed specification and topology is described on `CSIT LF -testbed wiki page <https://wiki.fd.io/view/CSIT/CSIT_LF_testbed>`_. +Detailed LF FD.io test bed specification and topology is described in +`wiki CSIT LF testbed <https://wiki.fd.io/view/CSIT/CSIT_LF_testbed>`_. -Testpmd Performance Tests Overview ----------------------------------- +Testing Summary +--------------- -Performance tests are split into two main categories: +Performance tests are split into the two main categories: - Throughput discovery - discovery of packet forwarding rate using binary search - in accordance to RFC2544. + in accordance with RFC2544. - - NDR - discovery of Non Drop Rate, zero packet loss. - - PDR - discovery of Partial Drop Rate, with specified non-zero packet loss. + - NDR - discovery of Non Drop Rate packet throughput, at zero packet loss; + followed by packet one-way latency measurements at 10%, 50% and 100% of + discovered NDR throughput. + - PDR - discovery of Partial Drop Rate, with specified non-zero packet loss + currently set to 0.5%; followed by packet one-way latency measurements at + 100% of discovered PDR throughput. - Throughput verification - verification of packet forwarding rate against - previously discovered throughput rate. These tests are currently done against + previously discovered NDR throughput. These tests are currently done against 0.9 of reference NDR, with reference rates updated periodically. -CSIT |release| includes following performance test suites: +CSIT |release| includes following performance test suites, listed per NIC type: - 2port10GE X520-DA2 Intel - - **L2XC** - L2 Cross-Connect forwarding of untagged Ethernet frames. + - **L2IntLoop** - L2 Interface Loop forwarding any Ethernet frames between + two Interfaces. + +Execution of performance tests takes time, especially the throughput discovery +tests. Due to limited HW testbed resources available within FD.io labs hosted +by Linux Foundation, the number of tests for NICs other than X520 (a.k.a. +Niantic) has been limited to few baseline tests. Over time we expect the HW +testbed resources to grow, and will be adding complete set of performance +tests for all models of hardware to be executed regularly and(or) +continuously. |