lib/connections: Use adaptive write size for rate limited connections (fixes #8630) #8631
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Outdated suggestions cannot be applied.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.
This is a best guess of what might be wrong in the issue in question, and an attempt to fix it. I've verified it doesn't harm performance in my local tests, and done some prints on the write sizes to verify they are reasonable, but I can't verify that it solves the problem because I haven't reproduced the problem.
That said, previously we'd do 8 KiB writes, which for 100 MB/s which is 12800 writes per second -- a lot of syscalls, and less opportunity for the TCP stack to send large segments maybe. Now we up the size to much larger for high rates, letting the lower layers do their thing in peace. This should improve things.