somu-imply opened a new pull request, #12439:
URL: https://github.com/apache/druid/pull/12439
The `latest` aggregator is not vectorized. This causes queries to take up a
lot of time while running this aggregator. This is the first part of a work in
progress where the longLast, floatLast and doubleLast aggregators are
vectorized. The three have been added to the benchmark tests as well where
there has been a 50% reduction in time.
```
Benchmark (query) (rowsPerSegment) (vectorize)
Mode Cnt Score Error Units
SqlExpressionBenchmark.querySql 38 5000000 false
avgt 5 176.557 ± 9.317 ms/op
SqlExpressionBenchmark.querySql 38 5000000 force
avgt 5 95.854 ± 3.002 ms/op
SqlExpressionBenchmark.querySql 39 5000000 false
avgt 5 139.142 ± 7.784 ms/op
SqlExpressionBenchmark.querySql 39 5000000 force
avgt 5 65.736 ± 5.199 ms/op
SqlExpressionBenchmark.querySql 40 5000000 false
avgt 5 129.723 ± 43.386 ms/op
SqlExpressionBenchmark.querySql 40 5000000 force
avgt 5 51.946 ± 4.960 ms/op
```
where
```
38-->LongLast
39-->DoubleLast
40-->FloatLast
```
This is a **work under progress as of now**.
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.
- [ ] 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]
