Auto-tools/Projects/Signal From Noise/Meetings/2013-02-28

From MozillaWiki
Jump to: navigation, search

Previous Action Items

  • [all] track the hanging ingestion job - is this finally resolved?
    • resolved
  • [all, future] come up with a solution for retriggers
  • [jeads] outline possible corner cases where we need a solution (i.e. thresholds)
    • [jmaher, all] work with perf team to solve this, ideally per page
  • [all, future] start verifying other data, tsvg*
    • future data to verify would be ts* and counters
  • [all] turn jobs orange for tp5n



  • concern about tp5 from John Daggett, specifically about font-loading/caching and how slowing down font rendering doesn't affect tp5 numbers.
    • considering making a tFont metric
  • bug 845943 - tscroll is not very useful
    • considering adding a new useful test and possibly rebuilding tscroll to be more accurate
  • tp5o is on the branches and waiting for the buildbot-configs to deploy (landed the buildbot-config change this morning)

round table from last week

  • BenWa has proposed adding thresholds per test to the tree, everything outside the threshold would be orange and require a perf sherriff to adjust the value. (carry over)
  • [tp5o] - Consider running only test pages which are stable, so we can turn datazilla on and depend on it
  • we should still consider looking at the hardware pool
    • Consider splitting hardware pool
    • Consider running talos on a few builds on a set of machines over and over to see what noise metrics we get (see how much noise is based on differences in machines versus in the build and install of the product)

Round Table

Action Items

  • [j*] tp5o analysis
  • [jeads] compare pgo vs non pgo
    • no pgo on OSX, only linux and windows
    • priority windows, p2 linux
    • p1 ts, p2 tp5n
  • [jeads] add machine names to the 1000 graphs