-
Notifications
You must be signed in to change notification settings - Fork 6.4k
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
Update protection info on recovered logs data #9875
Conversation
Summary: Implement key-value checksums on data recovered from the WAL. Test Plan: TBD Reviewers: Andrew Kryczka Subscribers: Tasks: T112036080 Tags: WAL
Summary: Changes originated in suggestions from ajkr. Test Plan: TBD Reviewers: Andrew Kryczka Subscribers: Tasks: T112036080 Tags: WAL
Added my own benchmark to description. I think the regression (7.2% on in-memory WAL) should be OK to tolerate without a flag for this feature considering it's crash-recovery only. Let me know if anyone disagrees. |
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.
IMO it'd be good to try the alternative approach suggested for handling prot_info_idx_ updates and see if it's cleaner. Otherwise LGTM.
@akomurav has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
@akomurav has updated the pull request. You must reimport the pull request before landing. |
@akomurav has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
@akomurav has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
1 similar comment
@akomurav has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
Update protection info on recovered logs data
Test Plan:
TEST_TMPDIR=/dev/shm/100MB_WAL_DB/ ./db_bench -benchmarks=fillrandom -write_buffer_size=1048576000
TEST_TMPDIR=/dev/shm/100MB_WAL_DB/ /usr/bin/time ./db_bench -use_existing_db=true -benchmarks=overwrite -write_buffer_size=1048576000 -writes=1 -report_open_timing=true
Mean regressed 7.2% (2201.4 -> 2359.9)