[
https://issues.apache.org/jira/browse/HBASE-14489?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14907666#comment-14907666
]
Andrew Purtell commented on HBASE-14489:
----------------------------------------
I'll look at the patch when at the desktop tomorrow.
I do remember microbenchmarking the classloader checks when we added them. You
can find the results on the jira that confirm your suspicions. It was a
necessary change but then hook calls were no longer 'free'. Even a coprocessor
that does nothing per hook will accrue that overhead and early out won't work
because something is installed.
> postScannerFilterRow consumes a lot of CPU
> ------------------------------------------
>
> Key: HBASE-14489
> URL: https://issues.apache.org/jira/browse/HBASE-14489
> Project: HBase
> Issue Type: Bug
> Reporter: Lars Hofhansl
> Labels: performance
> Attachments: 14489-0.98.txt
>
>
> During an unrelated test I found that when scanning a tall table with CQ only
> and filtering most results at the server, 50%(!) of time is spend in
> postScannerFilterRow, even though the coprocessor does nothing in that hook.
> We need to find a way not to call this hook when not needed, or to question
> why we have this hook at all.
> I think [~ram_krish] added the hook (or maybe [~anoop.hbase]). I am also not
> sure whether Phoenix uses this hook ([~giacomotaylor]?)
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
