Raise iosqeAsyncThreshold #8252
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
plaintext benchmark can cause Netty io_uring to use ASYNC SQEs that are not nicely handled at OS level (creating TONS of kernel threads) ie 16384 connections / 28 event loops = 582 connections per event loop which can use SYNC SEQs, which exceed the 25 default ones. This PR fix this behaviour and force SYNC SQEs: see netty/netty-incubator-transport-io_uring#152 (comment) for more info
franz1981
added a commit
to franz1981/FrameworkBenchmarks
that referenced
this pull request
Jun 23, 2023
plaintext benchmark can cause Netty io_uring to use ASYNC SQEs that are not nicely handled at OS level (creating TONS of kernel threads) ie 16384 connections / 28 event loops = 582 connections per event loop which can use SYNC SEQs, which exceed the 25 default ones. This PR fix this behaviour and force SYNC SQEs: see netty/netty-incubator-transport-io_uring#152 (comment) for more info
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
plaintext benchmark can cause Netty io_uring to use ASYNC SQEs that are not nicely handled at OS level (creating TONS of kernel threads ie ring size per event loop -> 4096 * 28 ).
Some more info on this behaviour at axboe/liburing#420 and https://blog.cloudflare.com/missing-manuals-io_uring-worker-pool/ as well.
In our case, plaintext could use 16384 connections partitioned in 28 event loops = 582 connections per event loop
which exceed the 25 default ones and can cause ASYNC SQEs to be used
This PR fix this behaviour and force SYNC SQEs: see netty/netty-incubator-transport-io_uring#152 (comment) for more info