11 ms·
Oh, my point isn't about whether throttling is good or bad, just about how it's measured. Go's GC's throughput is particularly bad in comparison to OpenJDK's GC
by pron 8y ago
Oh, my point isn't about whether throttling is good or bad, just about how it's measured. Go's GC's throughput is particularly bad in comparison to OpenJDK's GCs, so talking just about its GC pauses (which do not include throttling, AFAIK) without mentioning throughput gives a false impression.
- shipilev 8y agoI agree with you and present even a stronger statement: talking about GC pauses without measuring the actual end-to-end latencies may give the false impression! The overall throughput might look good, but threads might experience bad local latencies all over the place anyway. In Shenandoah project, that is my recurrent argument for capping the pacing stalls and going to STW mode when bounded pacing did not help.