16:03:17 #startmeeting vswitchperf 16:03:17 Meeting started Wed Jan 16 16:03:17 2019 UTC. The chair is srao. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:03:17 Useful Commands: #action #agreed #help #info #idea #link #topic. 16:03:17 The meeting name has been set to 'vswitchperf' 16:04:01 #topic Development Update. 16:04:34 #info Al Morton 16:05:13 #info Timestamps to Log-files- Should we? 16:06:46 #info Adding timestamps doesn't hurt. Consolidation of logs can be taken in future. 16:07:08 #agree 16:07:20 #info Testapi Container running in Pod-12. 16:07:55 #info Add this config file: OPNFV_URL= http://10.10.120.22:8000/a 16:08:09 #info add this into config file: OPNFV_URL= http://10.10.120.22:8000/api/v1 16:08:28 #info Include this when running the test: --opnfvpod intel-pod12 16:11:06 #info Started to explore Cedric's XTesting - create ansible for Vsperf. Will update the team next week. 16:14:14 #info Started experimenting with newer versions of OVS-DPDK-VPP 16:14:36 #info Need to discuss gRPC & REST limits with Cedric. 16:15:17 #info DPDK 18.11, OVS 2.10, VPP 18.1.2 16:18:10 #info Hari: We have to remove these flags '--burst=64 -i --txqflags=0xf00 ', if we move to DPDK 18.11 16:20:38 #info: Support for libvirt. We have to move from qemu-system-x86_64 command to libvirt. 16:21:44 #info Why it helps to move to libvirt: (a) For short-duration we may not see any issues, however for longer durations there will be performance degradation. 16:21:49 #info qemu will eventually cause a few packets to drop with VPP 16:22:10 #info estimate of 3 weeks development time 16:23:00 #info Why it helps to move to libvirt: (b) CPU pinning would be relatively easier 16:23:04 #info if we move to libvirt, the XML will tell qemu how to start, handled much more efficiently 16:24:02 #info Christian - With DPDK 18.11 We may have to enable legacy-memory option (--legacy-mem) 16:24:05 #info DPDK 18.11 might have to add legacy memory option, otherwise will see drops 16:24:30 #info Hari: --burst=64 i is still included, whereas disable-hw-vlan is not 16:26:45 #chair acm 16:26:45 Current chairs: acm srao 16:29:56 #topic Tipsy presentation 16:57:17 #info Peter and Felician will supply slides for posting in the read-out section of our Wiki ?? 16:58:06 #info t4p4s is the P4 switch, pronounced "tapas" 16:58:55 #info IEEE JSAC Dec 2018 paper describes the tool 17:00:53 #info performance results for lagopus and t4p4s were quite low. 17:03:18 #endmeeting