[
https://issues.apache.org/jira/browse/CASSANDRA-9738?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14655012#comment-14655012
]
Robert Stupp commented on CASSANDRA-9738:
-----------------------------------------
[~kohlisankalp] ok :)
Here are two more cstar results using [this
profile|https://gist.github.com/snazy/b6c160c65001eb074784].
[blade_11_b|http://cstar.datastax.com/graph?stats=7f7265a2-3aee-11e5-b022-42010af0688f&metric=op_rate&operation=3_user&smoothing=1&show_aggregates=true&xmin=0&xmax=190.96&ymin=0&ymax=113547.5]
shows the same OHC having the same throughput as 2.2 but with less GC pauses ;
both are faster than 3.0 alone.
[bdplab|http://cstar.datastax.com/graph?stats=af344e3e-3af0-11e5-b379-42010af0688f&metric=op_rate&operation=3_user&smoothing=1&show_aggregates=true&xmin=0&xmax=254.65&ymin=0&ymax=97972.6]
shows OHC having better throughput than 2.2 and 3.0 ; also less GC effort.
Both environments show this behaviour in every test - guess, it's ok to blame
it spinning disks and less RAM.
> Migrate key-cache to be fully off-heap
> --------------------------------------
>
> Key: CASSANDRA-9738
> URL: https://issues.apache.org/jira/browse/CASSANDRA-9738
> Project: Cassandra
> Issue Type: Sub-task
> Reporter: Robert Stupp
> Assignee: Robert Stupp
> Fix For: 3.0.0 rc1
>
>
> Key cache still uses a concurrent map on-heap. This could go to off-heap and
> feels doable now after CASSANDRA-8099.
> Evaluation should be done in advance based on a POC to prove that pure
> off-heap counter cache buys a performance and/or gc-pressure improvement.
> In theory, elimination of on-heap management of the map should buy us some
> benefit.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
