-
Notifications
You must be signed in to change notification settings - Fork 726
Support multithreaded benchmarks #2192
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
d144fa4
to
30165b2
Compare
Benchmark resultsInstruction countsSignificant differencesThere are no significant instruction count differences Other differencesClick to expand
Wall-timeSignificant differencesThere are no significant wall-time differences Other differencesClick to expand
Additional informationCheckout details:
|
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #2192 +/- ##
=======================================
Coverage 94.63% 94.63%
=======================================
Files 102 102
Lines 23928 23928
=======================================
Hits 22645 22645
Misses 1283 1283 ☔ View full report in Codecov by Sentry. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Nice :-)
ad3bdb3
to
fd6c77e
Compare
fd6c77e
to
bb784b5
Compare
bb784b5
to
The default client session store keeps 8 tickets per server. In this performance benchmark, we only talk to one server, so at thread counts higher than 8, we can starve this cache before it is refilled by the next successful connection. This means a full handshake would happen, which is undesirable in a test that wants to measure resumption speed (and, also, it fails an assertion.) So: allow any neccessary full handshakes to happen and simply do not include their costs in the collected timings.
d942e3a
to
a224b99
Compare
This produces output like:
This means:
old output
number of threads used
individual thread measurements
sum and average of thread measurements