#opnfv-vswitchperf: vswitchperf
Meeting started by SridharRao at 15:00:14 UTC
(full logs).
Meeting summary
- Development update (SridharRao, 15:02:43)
- Frame Size as IMIX-code in result
output. (SridharRao,
15:04:46)
- Question: Should We include writing results to
DB on Pod12. (SridharRao,
15:07:49)
- Reuse TestAPI framework (SridharRao,
15:11:53)
- Approach TSC with the issue and try to get it
resolved (SridharRao,
15:14:44)
- Mail to Tim@Ericsson (SridharRao,
15:15:29)
- Bob has uploaded the details of the Testing for
the Xena Patch. (SridharRao,
15:18:22)
- Pod12 and Pod19 (SridharRao, 15:19:07)
- Plugfest TestResults. (SridharRao, 15:19:36)
- In P2P - Not much transients and corresponding
effects. (SridharRao,
15:21:05)
- in PVP and PVVP - we see effects of the
transient processes - affecting the performance. (SridharRao,
15:21:30)
- Possible sources - Logical Interfaces and
VMs. (SridharRao,
15:21:53)
- Repeating had not much impact in P2P
case. (SridharRao,
15:23:40)
- for PVP and PVVP Throughput is almost twice as
high. Also, the many 'reversals' - repeating improved the
performance (zero loss) (SridharRao,
15:24:30)
- Summary: LV helps in improving
throughput. (SridharRao,
15:29:46)
- Pod12 and Pod19 (SridharRao, 15:45:04)
- Common Pod for Performance Tests? (SridharRao,
15:49:37)
- Testing projects should come up better usage
model of Pods - Explore reuse (SridharRao,
15:55:42)
Meeting ended at 15:59:44 UTC
(full logs).
Action items
- (none)
People present (lines said)
- SridharRao (20)
- collabot_ (3)
Generated by MeetBot 0.1.4.