Further improve memory usage of needle_map.CompactMap().
#6825
Conversation
This file contains hidden or 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
Since `CompactMap()` now uses variable-size slices as ID buckets, when needles are mostly ordered their capacity these buckets end up mostly filled, but with a lot of wasted cappacity. For the current bucket size of 10,000 elements, Golang allocates a capacity of 11,264, which adds up to ~12% waste. This MR ensures that buckets which end up filled are re-sized so their capacity matches size; on average, it improves memory usage by ~10% for higly ordered needle IDs. A couple `CompactMap()` memory pprofs yielded these averages: ``` v3.87, running weed server: 346.89MB/639.85MB heap, 347.16MB/3.48GB allocs v3.87, running weed volume: 346.89MB/626.50MB heap v3.88, running weed server: 468.81MB/751.42MB heap, 1.74GB/4.77GB allocs v3.88, running weed volume: 468.81MB/764.41MB heap new, running weed server: 426.21MB/711.50MB heap, 2.16GB/5.20GB allocs new, running weed volume: 426.21MB/664.94MB heap ``` Follow up to 2e1506c; see also seaweedfs#6804 .
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Outdated suggestions cannot be applied.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.
What problem are we solving?
#6804
How are we solving the problem?
As of 2e1506c,
CompactMap()uses variable-size slices as ID buckets.An unexpected issue is that mostly-ordered needle writes causes the majority of buckets to end up filled, but with a lot of unused capacity - i.e. wasted memory. For the current bucket size of 10,000 elements, Golang allocates a capacity of 11,264, or ~12% waste.
This MR ensures that buckets which end up filled are re-sized so their capacity matches size; on average, it improves memory usage by ~10% for mostly ordered needle IDs. A couple
CompactMap()memory pprofs yielded these averages:The underlying logic for
CompactMap.Set()was also simplified in the process.How is the PR tested?
Every single existing unit and integration test should pass without issues; the existing API and behavior for CompactMap() is unchanged.
Checks