Gehel added a comment.

comparing 6h of GC logs from before the reduction in heap size with 2h40m of GC logs after the reduction, here are a few observations:

  • GC overhead increased from 6.74% to 8.12%
  • allocation rate increased from ~900mb/sec to 1.2gb/sec (allocation rate is the rate at which the application creates objects, so not directly related to the change, but explains the GC overhead increase)
  • the allocation rate is sufficiently different than what I noted earlier that I suspect I read it wrong the first time:
  • allocation rate over that period is ~70MB/s (but probably peaks much higher, the updater has a fairly spiky workload over short periods of time)

Allocating ~1gb/sec is definitely a high rate! We probably need to start taking out a profiler and see if we can optimize blazegraph. We might be able to gain some throughput by further reducing the heap size or doing magical GC tuning, but we need to work on reducing the allocation rate as well.


TASK DETAIL
https://phabricator.wikimedia.org/T175919

EMAIL PREFERENCES
https://phabricator.wikimedia.org/settings/panel/emailpreferences/

To: Gehel
Cc: Stashbot, gerritbot, Smalyshev, Gehel, Aklapper, Lordiis, Lucas_Werkmeister_WMDE, GoranSMilovanovic, Adik2382, Th3d3v1ls, Ramalepe, Liugev6, QZanden, EBjune, merbst, Avner, Lewizho99, Maathavan, debt, Jonas, FloNight, Xmlizer, Izno, jkroll, Wikidata-bugs, Jdouglas, aude, Tobias1984, Manybubbles, Mbch331
_______________________________________________
Wikidata-bugs mailing list
Wikidata-bugs@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wikidata-bugs

Reply via email to