kfaraz opened a new pull request, #17390:
URL: https://github.com/apache/druid/pull/17390
### Description
In clusters with a large number of datasources, task operations block each
other owing to the `giant` lock in the `TaskLockbox`. This can become
particularly problematic in cases of streaming ingestion being done into
multiple datasources, where segment allocations become very slow since all of
them must pass through a single queue.
None of the task operations share a critical section across datasources and
it should suffice to perform the locking at the datasource level.
This patch is an attempt to remediate this problem.
### Changes
- Dedicate `TaskLockbox` to a single datasource.
- Add a `GlobalTaskLockbox` which delegates to the respective datasource for
any operation.
- The `GlobalTaskLockbox.sync()` must still be mutually exclusive from any
operation being performed on any of the datasources. This is performed by using
a `ReentrantReadWriteLock`.
### Pending
- Maintain `SegmentAllocationQueue` at datasource level.
- Fix up tests
- Clean up the `TaskLockbox` for a datasource when it is not needed anymore
- Try out changes in a cluster
---
This PR has:
- [ ] been self-reviewed.
- [ ] using the [concurrency
checklist](https://github.com/apache/druid/blob/master/dev/code-review/concurrency.md)
(Remove this item if the PR doesn't have any relation to concurrency.)
- [ ] added documentation for new or modified features or behaviors.
- [ ] a release note entry in the PR description.
- [ ] added Javadocs for most classes and all non-trivial methods. Linked
related entities via Javadoc links.
- [ ] added or updated version, license, or notice information in
[licenses.yaml](https://github.com/apache/druid/blob/master/dev/license.md)
- [ ] added comments explaining the "why" and the intent of the code
wherever would not be obvious for an unfamiliar reader.
- [ ] added unit tests or modified existing tests to cover new code paths,
ensuring the threshold for [code
coverage](https://github.com/apache/druid/blob/master/dev/code-review/code-coverage.md)
is met.
- [ ] added integration tests.
- [ ] been tested in a test Druid cluster.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
