Skip to content
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

Allow L0 -> L1 compaction use IOPriority::IO_MID #9918

Open
lhsoft opened this issue Apr 28, 2022 · 3 comments · May be fixed by #9999
Open

Allow L0 -> L1 compaction use IOPriority::IO_MID #9918

lhsoft opened this issue Apr 28, 2022 · 3 comments · May be fixed by #9999
Assignees

Comments

@lhsoft
Copy link
Contributor

lhsoft commented Apr 28, 2022

As we know that the slow L0 -> L1 compaction will cause write stall which cause high latency for online application.
Rocksdb have added IOPriority::IO_MID for rate limiter. It's better to all L0-L1 compaction use IO_MID to reduce the writing stall.

Expected behavior

Use can config the Rocksdb to use IO_MID for L0 -> L1 compaction for rate limiter.

Actual behavior

Steps to reproduce the behavior

@anand1976
Copy link
Contributor

CC @ajkr

@ajkr
Copy link
Contributor

ajkr commented Apr 28, 2022

Agreed. Also L0->L0 should be IO_MID. @lhsoft do you want to try it?

@lhsoft
Copy link
Contributor Author

lhsoft commented May 7, 2022

@ajkr sure, I'd like to work on it. Please feel free to assign it to me:)

@ajkr ajkr removed the up-for-grabs Up for grabs label May 8, 2022
@lhsoft lhsoft linked a pull request May 16, 2022 that will close this issue
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging a pull request may close this issue.

3 participants