-
Beta Was this translation helpful? Give feedback.
Replies: 2 comments 31 replies
-
NOTE: I can provide the flamegraph .svg files in case there is any interest. |
Beta Was this translation helpful? Give feedback.
-
That's how it often goes, CPUs are pretty good at copying small buffers, assuming it's <=1500B per request in your test, and zerocopy also adds some additional overhead. The kernel's networking stack is also pretty heavy in general, saving on small copies wouldn't make a drastic change. You might try to enable gso (e.g.
Yes, that's where it copies data.
That's interesting, I don't see immediately why. Can you share the svg of the profiles as you offered? |
Beta Was this translation helpful? Give feedback.
I've repeated the test to be longer than 1 minute order to remove the possible noises.
NOTES:
Operation not supported
as response toio_uring_register_ring_fd()
.As you can see there is now a 2% difference between
ZC
andNO_ZC
, beingZC
more efficient.ZC
NO ZC
Probably now we are seeing the benefits of using
ZC
compared to not using it, for our use case (small buffers). I hope everything makes sense.