dmitryor opened a new pull request #17096:
URL: https://github.com/apache/beam/pull/17096
Beam has two KafkaIO source implementations now:
* a modern one implemented as a Splittable DoFn (SDF), and
* a (deprecated) legacy one implemented as an SDF wrapper over an
UnboundedSource and KafkaUnboundedReader classes.
O found that the legacy KafkaIO source can not provide good throughput when
the latency of calls to Kafka
[Consumer.poll|https://github.com/apache/beam/blob/master/sdks/java/io/kafka/src/main/java/org/apache/beam/sdk/io/kafka/KafkaUnboundedReader.java#L523]
becomes high. The degradation is very sharp: a pipeline that drops elements
immediately after reading them from source was only able to read about 100-1000
qps per Kafka partition. The Kafka cluster was overprovisioned but was in a
remote network and had poll latency about 30ms.
The cause of throughput degradation is poor choice of the
[RECORDS_DEQUEUE_POLL_TIMEOUTT|https://github.com/apache/beam/blob/master/sdks/java/io/kafka/src/main/java/org/apache/beam/sdk/io/kafka/KafkaUnboundedReader.java#L334-L335]
set too low, it should be longer than any reasonable expected Kafka poll
latency. This is what's happening otherwise:
* consumerPollLoop fetches data bundle from Kafka via poll() and offers it
to the avaliableRecordsQueue
* message processing loop fetches bundle from avaliableRecordsQueue and
unblocks the consumerPollLoop
* consumerPollLoop calls poll() again
* message processing loop completes processing the bundle BEFORE the poll()
call above completes, and tries to fetch next bundle from avaliableRecordsQueue.
* fetch from avaliableRecordsQueue has a very short timeout (10ms) and if it
expires before the pending poll() in the consumerPollLoop completes the message
processing loop will believe there's no fresh data in Kafka and exit. All the
time until the message processing loop is rescheduled is wasted.
R: @lukecwik
------------------------
- [ ] [**Choose
reviewer(s)**](https://beam.apache.org/contribute/#make-your-change) and
mention them in a comment (`R: @username`).
- [ ] Format the pull request title like `[BEAM-XXX] Fixes bug in
ApproximateQuantiles`, where you replace `BEAM-XXX` with the appropriate JIRA
issue, if applicable. This will automatically link the pull request to the
issue.
- [ ] Update `CHANGES.md` with noteworthy changes.
- [ ] If this contribution is large, please file an Apache [Individual
Contributor License Agreement](https://www.apache.org/licenses/icla.pdf).
See the [Contributor Guide](https://beam.apache.org/contribute) for more
tips on [how to make review process
smoother](https://beam.apache.org/contribute/#make-reviewers-job-easier).
To check the build health, please visit
[https://github.com/apache/beam/blob/master/.test-infra/BUILD_STATUS.md](https://github.com/apache/beam/blob/master/.test-infra/BUILD_STATUS.md)
GitHub Actions Tests Status (on master branch)
------------------------------------------------------------------------------------------------
[](https://github.com/apache/beam/actions?query=workflow%3A%22Build+python+source+distribution+and+wheels%22+branch%3Amaster+event%3Aschedule)
[](https://github.com/apache/beam/actions?query=workflow%3A%22Python+Tests%22+branch%3Amaster+event%3Aschedule)
[](https://github.com/apache/beam/actions?query=workflow%3A%22Java+Tests%22+branch%3Amaster+event%3Aschedule)
See [CI.md](https://github.com/apache/beam/blob/master/CI.md) for more
information about GitHub Actions CI.
--
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]
