perf(arrow): move partition key into group map instead of cloning per row - #3159
Open
anoopj wants to merge 1 commit into
Open
perf(arrow): move partition key into group map instead of cloning per row#3159anoopj wants to merge 1 commit into
anoopj wants to merge 1 commit into
Conversation
… row RecordBatchPartitionSplitter::split groups a batch's rows by partition value into a HashMap. It iterated the partition Structs by reference and cloned each one into the entry key, allocating a fresh Struct (and, for string partitions, copying the string) for every row even though partition_structs is never used after the loop. Consume partition_structs with into_iter() and move each Struct into the entry, dropping the per-row clone. Behavior is unchanged: grouping is by key equality, unaffected by whether the key is moved or cloned. On a 100k-row string-partitioned batch, split() drops from ~430 to ~396 ns/row (~8%); numeric partitions see a smaller win.
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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 changes are included in this PR?
RecordBatchPartitionSplitter::split groups a batch's rows by partition value into a HashMap. It iterated the partition Structs by reference and cloned each one into the entry key, allocating a fresh Struct (and, for string partitions, copying the string) for every row even though partition_structs is never used after the loop.
Consume partition_structs with into_iter() and move each Struct into the entry, dropping the per-row clone.
Behavior is unchanged: grouping is by key equality, unaffected by whether the key is moved or cloned. On a 100k-row string-partitioned batch, split() drops from ~430 to ~396 ns/row (~8%); numeric partitions see a smaller win.
Are these changes tested?
Existing tests