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Outdated suggestions cannot be applied.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.
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
Add RangeDelAggregator microbenchmarks #4363
Add RangeDelAggregator microbenchmarks #4363
Changes from 1 commit
976f2c2
83dcbbb
6c64d6e
File filter
Filter by extension
Conversations
Jump to
There are no files selected for viewing
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.
Something to think about in the future. We might be getting artificially good cache perf by creating the tombstones right before they're added. While in practice, the range tombstone meta-block will often be cold when
AddTombstones()
is called.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.
I think at least some of the range tombstones (primarily the ones in the memtable and L0) will be warm if the user issues lots of reads, because those range tombstones will be read for each of those calls. Also, even disregarding cache perf, I think the relative benchmark changes would still be useful to measure performance improvements, since the cache hit rate should be the same before and after the changes. Regardless, this probably requires some more thought, so I left a TODO to look at this more carefully later.