feat: improve read performance by 7x with prebuffer - #1709
Merged
Conversation
rtyler
approved these changes
Oct 9, 2023
rtyler
left a comment
rtyler
left a comment
Member
There was a problem hiding this comment.
There is a memory tradeoff, but I think upstream defaulting indicates it's well worth the tradeoff for a default behavior.
Going to approve, thanks for another solid improvement @ion-elgreco
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Description
Enable prebuffer in the pyarrow.dataset.ParquetFragmentScanOptions. Relevant PR in Arrow repo, where they changed it to be default behavior. However, this won't be the case for older versions for PyArrow, so we need to set it to True.:
It improves read speed by 6-7x on Azure in one dataset that I have.
Before:
1min 4s ± 3.48 s per loop (mean ± std. dev. of 7 runs, 1 loop each)
After:
8.99 s ± 786 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)
Related Issue(s)
Closes ##1569