perf(reader): Optionally pass data file size to reader through FileScanTasks to avoid stat() call#2175
Open
mbutrovich wants to merge 4 commits intoapache:mainfrom
Open
perf(reader): Optionally pass data file size to reader through FileScanTasks to avoid stat() call#2175mbutrovich wants to merge 4 commits intoapache:mainfrom
FileScanTasks to avoid stat() call#2175mbutrovich wants to merge 4 commits intoapache:mainfrom
Conversation
mbutrovich
changed the title
FileScanTasks to avoid stat() callFileScanTasks to avoid stat() call
blackmwk
reviewed
Feb 25, 2026
Contributor
blackmwk
left a comment
blackmwk
left a comment
There was a problem hiding this comment.
Thanks @mbutrovich for this pr!
| let delete_filter_rx = | ||
| delete_file_loader.load_deletes(&task.deletes, Arc::clone(&task.schema)); | ||
|
|
||
| let file_size = if task.file_size_in_bytes > 0 { |
Contributor
There was a problem hiding this comment.
In what case the file_size_in_bytes is not larger than 0?
| ) -> Vec<Option<String>> { | ||
| let tasks = Box::pin(futures::stream::iter( | ||
| vec
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Which issue does this PR close?
stat()is taking over 17% of the CPU time increate_record_batch_stream_builder:What changes are included in this PR?
FileScanTask. Iceberg Java does this by wrapping a reference to aDataFilein itsFileScanTask. In this case we're just cherry-picking some fields until we decide we need more.Are these changes tested?
Existing tests.