Closed
Bug 1867959
Opened 10 months ago
Closed 10 months ago
Fix passing of submetric_summary_method. This switches Speedometer 3 to actually use the mean instead of the mean as the summary statistic.
Categories
(Testing :: Raptor, defect, P2)
Testing
Raptor
Tracking
(firefox122 fixed)
RESOLVED
FIXED
122 Branch
Tracking | Status | |
---|---|---|
firefox122 | --- | fixed |
People
(Reporter: sparky, Assigned: sparky)
References
Details
Attachments
(1 file)
Currently the submetric_summary_method isn't being passed all the way through results.py/output.py. This bug is to fix that.
Assignee | ||
Comment 1•10 months ago
|
||
Updated•10 months ago
|
Assignee: nobody → gmierz2
Status: NEW → ASSIGNED
Pushed by gmierz2@outlook.com: https://hg.mozilla.org/integration/autoland/rev/2462e8026fd7 Fix passing of submetric_summary_method through results.py/output.py. r=jrmuizel DONTBUILD
Comment 3•10 months ago
|
||
bugherder |
Status: ASSIGNED → RESOLVED
Closed: 10 months ago
status-firefox122:
--- → fixed
Resolution: --- → FIXED
Target Milestone: --- → 122 Branch
Comment 4•9 months ago
|
||
As a result of this checkin, we are now actually using the mean as the summary statistic on Speedometer 3, instead of the median of its 10 iterations.
We now have numbers which more realistically match the numbers that Speedometer 3 shows the user at the end of a run.
However, we are also more sensitive to outliers, so the noise has increased.
Bug 1868437 fixed the worst outliers but we still have others.
Summary: Fix passing of submetric_summary_method → Fix passing of submetric_summary_method. This switches Speedometer 3 to actually use the mean instead of the mean as the summary statistic.
You need to log in
before you can comment on or make changes to this bug.
Description
•