[jira] [Commented] (HBASE-17016) Remove per-region latency histogram metrics

["Gary Helmling (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/HBASE-17016?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15634364#comment-15634364 ] 

Gary Helmling commented on HBASE-17016:
---------------------------------------

We could also look at moving to HdrHistogram for all of our histogram usage.  That would eliminate the tons of Counter instances we generate and outperformed our FastLongHistogram in my testing.

> Remove per-region latency histogram metrics
> -------------------------------------------
>
>                 Key: HBASE-17016
>                 URL: https://issues.apache.org/jira/browse/HBASE-17016
>             Project: HBase
>          Issue Type: Task
>            Reporter: Andrew Purtell
>             Fix For: 2.0.0, 1.4.0
>
>
> Follow up from HBASE-10656, where [~enis] says:
> {quote}
> the main problem is that we have A LOT of per-region metrics that are latency histograms. These latency histograms create many many Counter / LongAdder objects. We should get rid of per-region latencies and maybe look at reducing the per-region metric overhead.
> {quote}
> Let's consider removing the per-region latency histograms until we have a low impact (in time and space) accounting.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)