hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Shaneal Manek (Commented) (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-5533) Add more metrics to HBase
Date Fri, 16 Mar 2012 00:56:38 GMT

    [ https://issues.apache.org/jira/browse/HBASE-5533?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13230792#comment-13230792
] 

Shaneal Manek commented on HBASE-5533:
--------------------------------------

Re line length: Ah, sorry about that. I was using 120 - fixed and will use 80 in the future.

Assign and declare together: Changed there (and in several other places). Will assign with
declaration where possible in the future.

Re timing overhead: Fixed it to cache calls to System.currentTimeMillis where possible (see:
CURRENT_TICK). Unfortunately, there is no good way to convert between System.nanoTime and
System.currentTimeMillis since the nanoTime epoch is undefined. I've attached a benchmark
showing that the overhead for System.nanoTime() is ~20ns. At the rate we're calling it that
should be negligible.

Load: I couldn't measure a difference in performance with this code and without it. I've used
a similar technique to get latency distributions on a service handling >10K requests/second
with no trouble, fwiw.

TSDB/UI: Yep, TSDB should be able to interpret it as a standard metric (so you can see you
95th percentile latencies over time, for example). I've uploaded a screenshot so you can see
what it looks like in the UI.
                
> Add more metrics to HBase
> -------------------------
>
>                 Key: HBASE-5533
>                 URL: https://issues.apache.org/jira/browse/HBASE-5533
>             Project: HBase
>          Issue Type: Improvement
>    Affects Versions: 0.92.2, 0.94.0
>            Reporter: Shaneal Manek
>            Assignee: Shaneal Manek
>            Priority: Minor
>         Attachments: BlockingQueueContention.java, HBASE-5533-0.92-v4.patch, hbase-5533-0.92.patch,
hbase5533-0.92-v2.patch, hbase5533-0.92-v3.patch, histogram_web_ui.png
>
>
> To debub/monitor production clusters, there are some more metrics I wish I had available.
> In particular:
> - Although the average FS latencies are useful, a 'histogram' of recent latencies (90%
of reads completed in under 100ms, 99% in under 200ms, etc) would be more useful
> - Similar histograms of latencies on common operations (GET, PUT, DELETE) would be useful
> - Counting the number of accesses to each region to detect hotspotting
> - Exposing the current number of HLog files

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message