AmatyaAvadhanula opened a new pull request, #17336:
URL: https://github.com/apache/druid/pull/17336
### Description
Polling the druid_segments and druid_pendingSegments tables can be a
bottleneck for several operations in Druid and the load on the metadata store
increases at least linearly with the number of segments in the cluster.
This PR aims to maintain a central cache of active segments and relevant
pending segments on the Overlord and use this wherever possible.
### Advantages
- Faster segment allocation (`40x` in
SegmentAllocateActionTest#testBenchmark) -> Lower lag especially in situations
where data is being backfilled
- (TODO?) Faster segment syncing on coordinator -> Faster ingestion handoff
on clusters where polling millions of segments (with hundreds of columns) takes
several seconds
### Changes
- Cache segment timelines for used segments and pending segments for
intervals with active locks on the overlord.
- Use this info for segment allocation
- (TODO) Make all the calls from the Coordinator go through the Overlord
before altering druid_segments or druid_pendingSegments in order to have a
centralized cache
- (TODO?) Use this cache and poll segments incrementally on the Coodinator
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]
