Open nmburgan opened 4 years ago
Would the next STS release (~1st week of Feb.) be reasonable, or do we need some more time to set this up and start doing this with the following release (probably LTS, may be another STS)?
Scale testing in CI Soak test rc build(s) Potentially soak test the actual release
Update Winston to create the soak testing tickets.
@billclaytor will create an epic for Scale testing in CI (BQ overhaul, etc...) Figure out how the priority compares with PPC and agent sim.
Epic has been created: https://tickets.puppetlabs.com/browse/SLV-756
In order to provide confidence that our next release does not introduce any performance regressions, and to be able to measure any performance gains we introduce, we should aim to have as much perf testing complete before release as possible, including scale and soak.
SLV should be one of the teams signing off on each release to ensure we are sending quality products out the door. What data do we need to be confident in no regressions? What do we need to work on to be enable this work in the appropriate timeframe? What touch points with other teams are needed?