datafusion-contrib / datafusion-contrib/liquid-cache
Implement threadpool based io_uring
Nobody has claimed this yet.
- Dominant language
- Rust
- Stars
- 452
- Forks
- 51
- Avg merge
- 3h 10m
- Merged PRs (30d)
- 12
Description
The idea is to delegate the io requests to a dedicated thread pool, which performs the blocking io_uring operations, once the data is fetched, the thread pool notifies/wakes the issuing task.
To implement this, we first need to start a thread pool, create a message passing worker (some MPMC queues) to wait for other threads to pass in a new io task.
Within the thread pool, we need to manually manage the low level io_uring primitives, this including the submission queues, completion queues, whether to poll the kernel or to wait for kernel interrupt, whether to colalense the io operations etc.
Once we get the results from kernel, we return the io data through message passing back to the task that initially issue the io.
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
No files, tests, or entry points are named. Start by locating the Rust I/O and object-store entry points, then trace how tasks currently issue and receive I/O. Done means the described thread pool, message passing, io_uring queue handling, and task wake-up flow are implemented and verified.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- rust
- Domain
- operating-systems
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100