[
https://issues.apache.org/jira/browse/HDDS-15164?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18078244#comment-18078244
]
Jheng-Sing Chen commented on HDDS-15164:
----------------------------------------
Thanks @[~adoroszlai] for pointing this out. I was not aware of HDDS-10021.
I will keep the JMH dependency and benchmark source out of the PR. The
benchmark was only run locally to measure the SCMRatisRequest/SCMRatisResponse
codec before/after performance.
I will post the results in HDDS-14647 and close this issue.
> Add a JMH benchmark to measure SCMRatisRequest and SCMRatisResponse codec
> performance.
> --------------------------------------------------------------------------------------
>
> Key: HDDS-15164
> URL: https://issues.apache.org/jira/browse/HDDS-15164
> Project: Apache Ozone
> Issue Type: Task
> Reporter: Jheng-Sing Chen
> Assignee: Jheng-Sing Chen
> Priority: Major
>
> The benchmark should measure encode/decode throughput separately for request
> and response paths. It should pre-generate benchmark inputs with a fixed
> random seed so that the measured time focuses on codec
> serialization/deserialization rather than test data generation.
> Initial benchmark coverage:
> - SCMRatisRequest encode
> - SCMRatisRequest decode
> - SCMRatisResponse encode
> - SCMRatisResponse decode
> - PipelineID protobuf payload
> - Long payload
> - List<PipelineID> payload
> Benchmark configuration:
> - JMH 1.37
> - Throughput mode
> - 5 warmup iterations
> - 10 measurement iterations
> - 3 forks
> - 1000 pre-generated input objects
> This benchmark can be used to compare SCM Ratis codec performance before and
> after HDDS-14647.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
