kfaraz opened a new pull request, #20708:
URL: https://github.com/apache/druid/pull/20708
### Changes
- Use running average metrics for Kafka poll-idle-ratio and average
processing rate (records / second)
- The average is taken over the entire lifecycle of a task and is not
exponentially weighted on purpose.
- This makes it easy to reason about the metric, and is driven by the
dominant behavior in the task duration.
- The poll-idle-ratio is reported by the Kafka consumer but is averaged over
a 30s window by default. This often leads to volatile metric values and the
metric value at the exact time of the auto-scaling decision wins. In production
setups, this becomes a die roll scenario since even though the metric had been
consistently reporting a certain value, a momentary spike or dip would cause
auto-scaler to make a bad decision.
- The avg processing rate metric is currently not used heavily as it was
also found to be volatile. So it is also being converted to a running average
in this patch solely for monitoring purposes so that if it is found to be
stable, we may consider using it as an alternative to the poll-idle-ratio.
<hr>
<!-- Check the items by putting "x" in the brackets for the done things. Not
all of these items apply to every PR. Remove the items which are not done or
not relevant to the PR. None of the items from the checklist below are strictly
necessary, but it would be very helpful if you at least self-review the PR. -->
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]
