[PERF] Remove upfront buffer allocations for local CSV reader #3242
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Outdated suggestions cannot be applied.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.
The local CSV reader currently makes upfront buffer allocations (80 MiB for file slabs and 80 MiB for CSV buffers). This unnecessarily blows up the read time for small CSV files which don't use so many buffers.
Since the local CSV reader allocates additional buffers as needed, we can remove all upfront allocations without affecting anything else in the implementation of the reader. This speeds up reads of small files.
At the same time, I benchmarked the performance of the reader against the test case described in #3055 and found no consistent slowdown without upfront comparisons.