perf(ext/http): increase buffer size to 64kb #16077
Closed
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Outdated suggestions cannot be applied.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.
Towards #13608
The current buffer size is the bottleneck. This PR increases the buffer size to 64kb, even if the largest possible size for a TLS packet is
16 * 1024 + 256
, we're reading at a slower rate than the upload speed, so packets are being buffered,deno/ext/http/lib.rs
Line 851 in e2828ad
You can check that
chunk.len()
is usually greater than the max TLS packet size.The higher the buffer size the faster it'll be, 64kb seems like a reasonable default value.
On my computer, streaming a 2gb upload to disk, takes 2.4s with this update down from 4.5s.
Adding a
bufSize/highWaterMark
option toDeno.serveHttp
orhttpConn.nextRequest()
to change the default value would be a good next step.