2.2. CSIT Release Notes

2.2.1. Changes in CSIT rls1801_2

  1. Added VPP performance tests

    • Container Service Chain Topologies Orchestrated by K8s with VPP Memif
      • Added tests with VPP vswitch in container connecting a number of VPP- in-container service chain topologies with L2 Cross-Connect and L2 Bridge-Domain configurations, orchestrated by Kubernetes. Added following forwarding topologies: i) “Parallel” with packets flowing from NIC via VPP to container and back to VPP and NIC; ii) “Chained” (a.k.a. “Snake”) with packets flowing via VPP to container, back to VPP, to next container, back to VPP and so on until the last container in a chain, then back to VPP and NIC; iii) “Horizontal” with packets flowing via VPP to container, then via “horizontal” memif to next container, and so on until the last container, then back to VPP and NIC;
    • VPP TCP/IP stack
      • Added tests for VPP TCP/IP stack using VPP built-in HTTP server. WRK traffic generator is used as a client-side;
    • SRv6
      • Initial SRv6 (Segment Routing IPv6) tests verifying performance of IPv6 and SRH (Segment Routing Header) encapsulation, decapsulation, lookups and rewrites based on configured End and End.DX6 SRv6 egress functions;
    • IPSecSW
      • SW computed IPSec encryption with AES-GCM, CBC-SHA1 ciphers, in combination with IPv4 routed-forwarding;

  2. Presentation and Analytics Layer

    • Added throughput speedup analysis for multi-core and multi-thread VPP tests into Presentation and Analytics Layer (PAL) for automated CSIT test results analysis;

  3. Other changes

    • Framework optimizations
      • Ability to run CSIT framework on ARM architecture;
      • Overall stability improvements;
    • NDR and PDR throughput binary search change
      • Increased binary search resolution by reducing final step from 100kpps to 50kpps;
    • VPP plugin loaded as needed by tests
      • From this release only plugins required by tests are loaded at VPP initialization time. Previously all plugins were loaded for all tests;

2.2.2. Performance Changes

Relative performance changes in measured packet throughput in CSIT rls1801_2 are calculated against the results from CSIT rls1801 report. Listed mean and standard deviation values are computed based on a series of the same tests executed against respective VPP releases to verify test results repeatibility, with percentage change calculated for mean values. Note that the standard deviation is quite high for a small number of packet throughput tests, what indicates poor test results repeatability and makes the relative change of mean throughput value not fully representative for these tests. The root causes behind poor results repeatibility vary between the test cases.

2.2.2.1. NDR Throughput Changes

NDR small packet throughput changes between releases are available in a CSV and pretty ASCII formats:

2.2.2.2. PDR Throughput Changes

NDR small packet throughput changes between releases are available in a CSV and pretty ASCII formats:

Measured improvements are in line with VPP code optimizations listed in VPP-18.01 release notes.

2.2.3. Known Issues

Here is the list of known issues in CSIT rls1801_2 for VPP performance tests:

# Issue Jira ID Description
1 Vic1385 and Vic1227 low performance. VPP-664 Low NDR performance.
2 Sporadic (1 in 200) NDR discovery test failures on x520. CSIT-570 DPDK reporting rx-errors, indicating L1 issue. Suspected issue with HW combination of X710-X520 in LF testbeds. Not observed outside of LF testbeds.
3 Lower than expected NDR throughput with xl710 and x710 NICs, compared to x520 NICs. CSIT-571 Suspected NIC firmware or DPDK driver issue affecting NDR and PDR throughput. Applies to XL710 and X710 NICs.
4 QAT IPSec scale with 1000 tunnels (interfaces) in 2t2c config, all tests are failing. VPP-1121 VPP crashes during configuration of 1000 IPsec tunnels. 1t1c tests are not affected
5 rls1801 plugin related performance regression CSIT-925 With all plugins loaded NDR, PDR and MaxRates vary intermittently from 3% to 5% across multiple test executions. Requires plugin code bisecting.
6 rls1801 generic small performance regression ip4base, l2xcbase, l2bdbase CSIT-926 Generic performance regression of discovered NDR, PDR and MaxRates of -3%..-1% vs. rls1710, affects ip4base, l2xcbase, l2bdbase test suites. Not detected by CSIT performance trending scheme as it was masked out by another issue CSIT-925.
7 rls1801 substantial NDR performance regression for vhost-user vring size of 1024 CSIT-927 Much lower NDR for vhostvr1024 tests, with mean values regression of -17%..-42% vs. rls1710, but also very high standard deviation of up to 1.46 Mpps => poor repeatibility. Making mean values not fully representative.
8 rls1801 substantial NDR/PDR regression for IPSec tunnel scale with HW QAT crypto-dev CSIT-928 NDR regression of -7%..-15%, PDR regression of -3%..-15% compared to rls1710.
9 NAT plugin requires ACL plugin enabled in VPP release 18.01.2 N/A Because of nodes order on interface feature all tests related to NAT plugin needs to have enabled ACL plugin too.