14:45:07 #startmeeting OPNFV VSPERF 14:45:07 Meeting started Wed Aug 9 14:45:07 2017 UTC. The chair is trevor_intel. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:45:07 Useful Commands: #action #agreed #help #info #idea #link #topic. 14:45:07 The meeting name has been set to 'opnfv_vsperf' 14:45:17 #info Trevor Cooper 15:01:42 #info Al Morton 15:03:15 #topic 2. Release milestones update 15:05:25 #info Need to merge TREX integration and new TCs from Martin 15:06:51 #info patches for Trex integration and Martin's JIRA: VSPERF-518 should be completed by Aug 11 MS5 15:07:14 #info Any documentation updates? 15:07:21 #topic Documentation 15:07:39 #info Tuning POD work may add some documentation 15:08:38 #topic Ixia license 15:09:03 #info Mars used Ixia so it works 15:10:26 #topic Integration of SW version of IXNetwork - need input from Pierre (OOO this week) 15:10:27 #topic Ixia integration work 15:10:46 #info Still waiting for description of the work from Ixia 15:11:18 #topic Long Duration Tests 15:12:39 #info Test WG has discussed long duration tests, need Pod resources for this, logistics with 5 installers are complex 15:13:12 #info what are test cases we would like to include? Long duration means 2 weeks. 15:14:25 #info Trevor proposed latency test cases ... need to tune up what we collect to make them useful 15:14:39 #info trevor_intel proposes latency tests for long duration 15:17:37 #info Al suggests an alternative is many short duration tests ... repeated ... makes results duration easier 15:18:53 #info look at min max ave collected from many short duration latency tests 15:18:53 #info Sridhar ... what aspect of OVS should we focus on? 15:19:32 #info for example varying number of flows 15:25:25 #info Trevor proposes going with Al's suggestion to use current latency tests and repeat 15:25:53 #info follow-up proposal at Test WG meeting this week 15:26:02 #topic Test API 15:28:33 #info need to discuss naming of a few test cases 15:28:59 #info need to fix time display (currently UNIX format) 15:29:29 #info time for each test should be replrted seperately 15:30:27 #info summary, some small problems and still some things to do, Mars will supply links 15:30:54 #link https://wiki.opnfv.org/display/vsperf/VSPerf+testcases+report 15:32:37 #info Mars described this report on the Wiki - this is a list of all test cases VSPERF has 15:35:33 #info does not include ?? This is the list of tests included in the API (results collected and stored through the API??) 15:40:52 #action Trevor schedule a separate meeting to discuss including Morgan or Serena 15:41:42 #topic IMIX 15:41:56 #info JIRA- 521 15:43:37 #link https://jira.opnfv.org/browse/VSPERF-521 15:45:02 #info Martin and Al discuss 520 and 521 with a view to merge and assign tasks 15:46:43 #action Martin K and acm discuss 520 and 521 with a view to merge and assign tasks 15:47:47 #topic Sridhar's topics: Stress VMs 15:49:24 #info related to JIRA 523, Noisy Neighbor tests, so integration of Stress VMs is a sub-task of 523 15:51:15 #info Sridhar's VMs at github.com/opensource-tnbt/stressng-images 15:51:54 #info Sridhar created thiese VMs after discussion with Stress-ng author, Colin King 15:53:14 #action Sridhar will put the VM images into OPNFV artifacts 15:53:34 #info need to store the VMs with our other VMs, like VLOOP, but apparently need special write access, trouble ticket 15:54:07 #info 523 is a proposal for suffix to test case name 15:57:54 #info ACE Lee has some MoonGen related questions - should mail to the VSPERF list 15:58:03 #info Ace Lee would like to proposes fixes to Moongen ... suggest send suggestions and questions to VSPERF mailing list 15:59:14 #info SR's 524 - live results from traffic Generators 16:00:33 #info SR's 525 - want to include *collectd* these all support the bigger task of *correlation* 16:01:01 #info Need a bigger task or Story or Epic to group all these tasks. 16:01:39 #topic Latency histogram JIRA -522 16:03:34 #topic CI tuning 16:03:39 #topic CI tuning - Christian 16:04:28 #info has done much to isolate CPUs, want to try this profile on the Sandbox, ... 16:05:06 #info ... but this profile is for Centos, most of Pod uses Ubuntu 16.04 16:06:25 #info uses tuneadm profile, have a new one for Centos 7.3, if recommend this to the community then we shouldget similar numbers 16:08:20 #info so, documentation needed recommending a consistent baseline configuration and these are the perf numbers you should get... 16:09:31 #endmeeting