You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "ASF GitHub Bot (JIRA)" <ji...@apache.org> on 2017/07/13 15:32:00 UTC
[jira] [Commented] (FLINK-7175) Add simple benchmark suite for
Flink
[ https://issues.apache.org/jira/browse/FLINK-7175?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16085882#comment-16085882 ]
ASF GitHub Bot commented on FLINK-7175:
---------------------------------------
GitHub user pnowojski opened a pull request:
https://github.com/apache/flink/pull/4323
[FLINK-7175] Add first simplest Flink benchmark
Example output:
```
Benchmark (objectReuse) (parallelism) (stateBackend) Mode Cnt Score Error Units
EventCountBenchmark.benchmarkCount true 1 memory thrpt 5 4433.286 ± 777.338 ops/ms
EventCountBenchmark.benchmarkCount true 1 rocks thrpt 5 569.078 ± 15.298 ops/ms
```
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/pnowojski/flink benchmarks
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/flink/pull/4323.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #4323
----
commit af6ea42ecdac723cef36f9d994cac3c680114ddd
Author: Piotr Nowojski <pi...@gmail.com>
Date: 2017-07-05T11:39:20Z
[FLINK-7175] Add first simplest Flink benchmark
Example output:
Benchmark (objectReuse) (parallelism) (stateBackend) Mode Cnt Score Error Units
EventCountBenchmark.benchmarkCount true 1 memory thrpt 5 4433.286 ± 777.338 ops/ms
EventCountBenchmark.benchmarkCount true 1 rocks thrpt 5 569.078 ± 15.298 ops/ms
----
> Add simple benchmark suite for Flink
> ------------------------------------
>
> Key: FLINK-7175
> URL: https://issues.apache.org/jira/browse/FLINK-7175
> Project: Flink
> Issue Type: Improvement
> Reporter: Piotr Nowojski
> Assignee: Piotr Nowojski
>
> For a long term goal it would be great to have both full scale (with real clusters) and micro benchmarks suites that runs automatically against each PR and constantly against master branch.
> First step towards this is to implement some local micro benchmarks, that run some simple Flink applications on local Flink cluster. Developers could use those benchmarks manually to test their changes.
> After that, we could setup some simple automation tool that would run those benchmarks against master branch on some old computer/laptop. We could publish it to our instance of [Codespeed|https://github.com/tobami/codespeed] project (example instance for pypy project: [PyPy Speed Center|http://speed.pypy.org/])
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)