perf: apply projection when reading checkpoint parquet #2717
Conversation
This file contains 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
alexwilcoxson-rel
requested review from
wjones127,
roeap,
rtyler and
hntd187
as code owners
rtyler
approved these changes
Jul 30, 2024
github-merge-queue
bot
removed this pull request from the merge queue due to failed status checks
github-merge-queue
bot
removed this pull request from the merge queue due to failed status checks
github-merge-queue
bot
removed this pull request from the merge queue due to failed status checks
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Outdated suggestions cannot be applied.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.
Description
This change applies projection pushdown when reading the checkpoint parquet file. Typically only a portion of the schema is read. Adds + Txns, Removes, or Metadata+Protocol.
Note: I also attempted to apply a RowFilter. This works but provided no additional gain, and in the event every column was projected it slowed things down.
The results show that the old code is the same roughly as the new code when projecting every column. However when projecting just M+P actions there is a about a 2.75x improvement.
I have less data on this but the full EagerSnapshot on my 1,000,000 action table went from ~600-700ms to ~150ms.
Related Issue(s)
Documentation