[ 
https://issues.apache.org/jira/browse/CASSANDRA-5361?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13608255#comment-13608255
 ] 

Ryan McGuire commented on CASSANDRA-5361:
-----------------------------------------

I've run a 4-node baseline off trunk without +UseTLAB and here is the average 
result:

{code}
  stress -n 2000000 -o read -i 1

  Averages from the middle 80% of values:
  interval_op_rate          : 21918
  interval_key_rate         : 21918
  latency median            : 0.9
  latency 95th percentile   : 2.7
  latency 99.9th percentile : 40.2
  Total operation time      : 00:01:42
  END
{code}

With +UseTLAB I do indeed get better results:

{code}
  Averages from the middle 80% of values:
  interval_op_rate          : 28352
  interval_key_rate         : 28352
  latency median            : 0.7
  latency 95th percentile   : 1.4
  latency 99.9th percentile : 22.3
  Total operation time      : 00:01:21
  END
{code}

Looking at interval_op_rate, that's ~30% increase in read performance. 
Interestingly, I see more variation in the interval_op_rate with UseTLAB - the 
first run I ran was much closer to the baseline at 23065. I never saw that much 
variation without UseTLAB, but I never did see a value that was below the 
baseline average, so I see it as a benefit to turn it on even though it's less 
consistent.

Write performance did not noticeably change. 
                
> Enable ThreadLocal allocation in the JVM
> ----------------------------------------
>
>                 Key: CASSANDRA-5361
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-5361
>             Project: Cassandra
>          Issue Type: Improvement
>         Environment: JDK 6
>            Reporter: T Jake Luciani
>            Assignee: Ryan McGuire
>             Fix For: 1.2.4
>
>
> Adding -XX:+UseTLAB to cassandra-env.sh yields a ~15% read performance boost.
> Works with usual stress tool but also in our production.
> It enables thread local allocation of new objects which makes sense for SEDA 
> based systems.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Reply via email to