-
Notifications
You must be signed in to change notification settings - Fork 969
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
perf(shwap/bitswap): add session pools inside Getter #3947
Conversation
In Robusta enviroment this change: * Improved load distribution across BN/FNs * Sped up sampling Originally, I was testing Bitswap with this change, but decided agaisnt including it in the patch, as its solely an optimization. However, without this change load distribution was rather strange. One LN sampled at 180/s, while all other at 4/s. Adding pools seemed to fix the issue.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
SGTM
I didn't realize that @walldiss is gonna take over that. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks @walldiss. LGTM
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
approving on behalf of @Wondertan
In the Robusta environment this change: * Improved load distribution across BN/FNs * Sped up sampling Originally, I was testing Bitswap with this change but decided against including it in the patch, as it is solely an optimization. However, without this change on the rc tag, load distribution was rather strange. One LN sampled at 180/s, while all 4 others at 4/s. Adding pooling fixed the issue and resulted in all 5 LNs at about 100/s. Co-authored-by: Vlad <[email protected]>
In the Robusta environment this change:
Originally, I was testing Bitswap with this change but decided against including it in the patch, as it is solely an optimization. However, without this change on the rc tag, load distribution was rather strange. One LN sampled at 180/s, while all 4 others at 4/s. Adding pooling fixed the issue and resulted in all 5 LNs at about 100/s.