You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@kafka.apache.org by "Ismael Juma (JIRA)" <ji...@apache.org> on 2016/05/11 02:21:12 UTC

[jira] [Commented] (KAFKA-3695) Make sure benchmarks run for long enough for reliable results

    [ https://issues.apache.org/jira/browse/KAFKA-3695?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15279394#comment-15279394 ] 

Ismael Juma commented on KAFKA-3695:
------------------------------------

cc [~apovzner]

> Make sure benchmarks run for long enough for reliable results
> -------------------------------------------------------------
>
>                 Key: KAFKA-3695
>                 URL: https://issues.apache.org/jira/browse/KAFKA-3695
>             Project: Kafka
>          Issue Type: Improvement
>            Reporter: Ismael Juma
>
> The benchmarks in `benchmark_test.py` with the default settings run for a relatively short period of time (between 1s and 2s is common last time I tested).
> [~becket_qin] suggested that the test should be running for a lot longer than that to ensure reliable results. Something between 10s and 15s would probably be a good target.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)