You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Brandon Williams (JIRA)" <ji...@apache.org> on 2009/10/26 20:38:59 UTC
[jira] Commented: (CASSANDRA-514) stress.py fixes and improvements
[ https://issues.apache.org/jira/browse/CASSANDRA-514?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12770156#action_12770156 ]
Brandon Williams commented on CASSANDRA-514:
--------------------------------------------
I should also note that I changed STDEV to be a function of the TOTAL_KEYS. At 30%, there's around a 60-68% repeat rate on reads.
> stress.py fixes and improvements
> --------------------------------
>
> Key: CASSANDRA-514
> URL: https://issues.apache.org/jira/browse/CASSANDRA-514
> Project: Cassandra
> Issue Type: Improvement
> Environment: debian amd64 Java(TM) SE Runtime Environment (build 1.6.0_12-b04)
> Reporter: Brandon Williams
> Priority: Minor
> Attachments: multiproc.patch
>
>
> stress.py cannot currently fully stress cassandra. This is due to at least two reasons:
> 1. It does not use the accelerated thrift protocol (thanks jbellis)
> 2. The GIL is preventing concurrency
> To address this, I've modified it in a backwards-compatible way to support multiprocessing, if available.
> Also, key_generator was providing floats instead of integers, causing the read test to ask for keys that did not exist.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.