[
https://issues.apache.org/jira/browse/CASSANDRA-9558?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Andy Tolbert updated CASSANDRA-9558:
------------------------------------
Attachment: CASSANDRA-9558-ProtocolV2.patch
Attached [^CASSANDRA-9558-ProtocolV2.patch] which forces protocol version 2 in
the stress tests. Alternatively we could use a separate driver jar for stress.
Obviously this is not ideal, but it will force the java-driver to use a dynamic
pool size implementation with 2 core and 8 max connections. After applying
write ops went from 90274/s to 141773/s which seems more in line with the
results expected based on the results in the description. I'll log a
java-driver issue to allow dynamic pool sizing with protocol version 3+ as
these results demonstrate that using only 1 connection per host has a
noticeable impact.
> Cassandra-stress regression in 2.2
> ----------------------------------
>
> Key: CASSANDRA-9558
> URL: https://issues.apache.org/jira/browse/CASSANDRA-9558
> Project: Cassandra
> Issue Type: Bug
> Reporter: Alan Boudreault
> Priority: Blocker
> Attachments: 2.1.log, 2.2.log, CASSANDRA-9558-ProtocolV2.patch
>
>
> We are seeing some regression in performance when using cassandra-stress 2.2.
> You can see the difference at this url:
> http://riptano.github.io/cassandra_performance/graph_v5/graph.html?stats=stress_regression.json&metric=op_rate&operation=1_write&smoothing=1&show_aggregates=true&xmin=0&xmax=108.57&ymin=0&ymax=168147.1
> The cassandra version of the cluster doesn't seem to have any impact.
> //cc [~tjake] [~benedict]
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
