[
https://issues.apache.org/jira/browse/HBASE-13291?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-13291:
--------------------------
Attachment: q (1).png
Here is before and after (two runs were not for same amount of time). GC is
about the same (not posted). Looks like patch makes for about 6-7% more
throughput. Should probably commit.
Let me see if I can find a better place to go digging.
(This is 5 servers each running a process that hosts 10 clients doing PE
scan1000 against a zipfian dataset of ten columns of average size 180k or so)
> Lift the scan ceiling
> ---------------------
>
> Key: HBASE-13291
> URL: https://issues.apache.org/jira/browse/HBASE-13291
> Project: HBase
> Issue Type: Improvement
> Components: Scanners
> Affects Versions: 1.0.0
> Reporter: stack
> Assignee: stack
> Attachments: 13291.inlining.txt, q (1).png, traces.7.svg,
> traces.filterall.svg, traces.nofilter.svg, traces.smaller.svg
>
>
> Scanning medium sized rows with multiple concurrent scanners exhibits
> interesting 'ceiling' properties. A server runs at about 6.7k ops a second
> using 450% of possible 1600% of CPUs when 4 clients each with 10 threads
> doing scan 1000 rows. If I add '--filterAll' argument (do not return
> results), then we run at 1450% of possible 1600% possible but we do 8k ops a
> second.
> Let me attach flame graphs for two cases. Unfortunately, there is some
> frustrating dark art going on. Let me try figure it... Filing issue in
> meantime to keep score in.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
