Reuse file descriptors for LocalFileSystem object storage
- Dominant language
- Rust
- Stars
- 9.3k
- Forks
- 2.4k
- Avg merge
- 3d 7h
- Merged PRs (30d)
- 344
Description
### Is your feature request related to a problem or challenge?
The ObjectStore implementation is stateless and doesn't allow for passing `File` descriptors.
This means for every request it's doing a `File::open` (and close) call, which creates overhead (mainly when doing a lot of small requests, e.g. with predicate pushdown enabled).
### Describe the solution you'd like
Implement an object store implementation for DataFusion that opens / reuses the `File`s for each path.
We can use `Arc` stored in hashmap for each file / path + read_at (stateless file read) to minimize overhead and maximize concurrency.
### Describe alternatives you've considered
_No response_
### Additional context
Also see
https://github.com/apache/arrow-rs-object-store/issues/622
https://github.com/apache/arrow-rs-object-store/issues/614
Contributor guide
Research direction
Start with DataFusion's ObjectStore and LocalFileSystem entry points, then review the linked arrow-rs-object-store issues #622 and #614 for related constraints. The work is complete when LocalFileSystem reuses open File descriptors per path while preserving concurrent stateless reads and avoiding an open/close cycle for every request.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- rust
- Domain
- backend, data-engineering, performance
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100