[
https://issues.apache.org/jira/browse/HBASE-19285?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16257841#comment-16257841
]
Andrew Purtell commented on HBASE-19285:
----------------------------------------
bq. Well, color me surprised. This seems to be "OK".
Like you say maybe it's the aggregation up to tables that keeps both the heap
utilization and perf impact greatly reduced from before. I didn't do any of the
prior measurement so unfortunately have no insight there.
bq. I think I could even be swayed to have them be default=false to remove all
possibility for unexpected perf impact :)
If there is not a significant impact defaulting to true should be fine. They
can be disabled if there is a problem. A default of false is going to surprise
people more I think. Earlier branch RMs may have a different opinion.
The 1.4 RC keeps getting pushed back but I am about to commit HBASE-19239 and
then have nothing else to do but spin binaries and exercise them. Will do that
after getting back from Thanksgiving holiday. If you'd like to see these go out
in 1.4 get a review and commit them before then and I'll also do a perf
assessment.
> Add per-table latency histograms
> --------------------------------
>
> Key: HBASE-19285
> URL: https://issues.apache.org/jira/browse/HBASE-19285
> Project: HBase
> Issue Type: Bug
> Components: metrics
> Reporter: Clay B.
> Assignee: Josh Elser
> Priority: Critical
> Fix For: 2.0.0, 1.4.0, 1.3.3
>
>
> HBASE-17017 removed the per-region latency histograms (e.g. Get, Put, Scan at
> p75, p85, etc)
> HBASE-15518 added some per-table metrics, but not the latency histograms.
> Given the previous conversations, it seems like it these per-table
> aggregations weren't intentionally omitted, just never re-implemented after
> the per-region removal. They're some really nice out-of-the-box metrics we
> can provide to our users/admins as long as it's not detrimental.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
