Gold-180 - fix: inproper nodejs task scheduling, use mpsc for multi_send #10
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
arhamj
reviewed
Jul 16, 2024
arhamj
reviewed
Jul 16, 2024
kgmyatthu
commented
Jul 26, 2024
mhanson-github
force-pushed
the
gold-180
branch
from
dfcb30d
to
0131136
Compare
The why - In multi send neon binding the function is spawning tokio threads for every ip:port pair along with O(n) oneshot channel pairs with nodejs callback function being Atomically reference counted this bring no benefit whatsoever while impact performance and lost thread yeilding and lost thread wakers on non-ideal/edge situation.
mhanson-github
force-pushed
the
gold-180
branch
from
0131136
to
b0b7cf0
Compare
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Outdated suggestions cannot be applied.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.
In multi send neon binding the function is spawning tokio
threads for every ip:port pair along with O(n) oneshot channel pairs
with nodejs taskqueue scheduling callbacks for O(n) n = target addresses. In addition we have O(n) spawn for actual connection outbound stuff (this part is legit). This make multi send spawn O(n^2) tokio threads.
This code is also having problem with failure to give control back to runtime thread and lost thread wakers on non-ideal/edge situation. This lead to zombie thread , tokio runtime thread not being able to get back resources for the during of the whole application cycle. (Node cycle).
This PR fixes this, by using MPSC channel, eliminate thread spawning unnecessarily in neon bindings, thus reducing thread spawn back to O(n).