[
https://issues.apache.org/jira/browse/CASSANDRA-14232?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16417847#comment-16417847
]
Jason Brown commented on CASSANDRA-14232:
-----------------------------------------
On the whole this is fine. However, I don't think we should include the CAS
write latencies with the general write latencies. CAS, by definition, will
involve several round trips, and that might make the metric confusing for
operators. You could move the CAS metrics to a separate 'coordinator write
latency' metric, but you could also drop it as we already have
{{StorageProxy#casWriteMetrics}}.
> Add metric for coordinator writes per column family
> ---------------------------------------------------
>
> Key: CASSANDRA-14232
> URL: https://issues.apache.org/jira/browse/CASSANDRA-14232
> Project: Cassandra
> Issue Type: Improvement
> Reporter: Sumanth Pasupuleti
> Assignee: Sumanth Pasupuleti
> Priority: Major
> Fix For: 4.0
>
> Attachments: 14232-trunk.txt
>
>
> Includes write ops and latencies at coordinator per column family.
> Relevant discussion in dev mailing list -
> [https://lists.apache.org/thread.html/f68f694b13b670a1fa28fa75620304603fc89e94ec515933199f4c37@%3Cdev.cassandra.apache.org%3E]
> Below are a few advantages of having such metric
> * Ability to identify specific column family that coordinator writes are
> slow to
> * Also useful in a multi-tenant cluster, where different column families are
> owned by different teams
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
