OmniaGM commented on PR #15968:
URL: https://github.com/apache/kafka/pull/15968#issuecomment-2821525320
> @OmniaGM : Thanks for the updated PR. A few more comments.
>
> Also, could you run producer performance test to make sure there is no
performance degradation with this PR? Thanks.
I have ran quick perf tests locally with 3 brokers (with 3 dedicated
controllers in kraft mode), topic has 10 partitions and 3 replicas
```
kafka-producer-perf-test.sh \
--producer-props bootstrap.servers=localhost:9092 linger.ms=100
batch.size=10000 acks=-1 \
--topic test-topic --num-records 5000000 --record-size 1024 --throughput -1
```
The output was the following
the MB/sec is more/ or less the same with close avg latency. There is some
diff in max latency and p99.99th
| measurement | with topic id | without topic id |
| --------------- | ------------ | ---------------- |
| records/sec. | 17968.9 | 17986.4 |
| MB/sec. | 17.55. | 17.56. |
| avg latency. | 1666.01ms | 1664.09 |
| max latency. | 10589.00ms| 9798.00ms |
| 50th. | 1516ms. | 1461ms |
| 95th, | 2216ms. | 2563ms |
| 99th, | 5134ms. | 5266ms |
| 99.9th. | 9182ms. | 8498ms |
--
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: jira-unsubscr...@kafka.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
