hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Lars Hofhansl (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-10047) postScannerFilterRow consumes a lot of CPU in tall table scans
Date Wed, 27 Nov 2013 18:29:35 GMT

    [ https://issues.apache.org/jira/browse/HBASE-10047?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13834013#comment-13834013
] 

Lars Hofhansl commented on HBASE-10047:
---------------------------------------

I think the costly part is the instanceof. So if a coprocessor could make its type and the
methods it overloads (as you said) explicit that could save some cycles.
However, coprocessors are not on the superhot code paths. The one mentioned here being an
exception as it could called a lot when we filter a lot of rows at the server. So I am not
sure it is actually worth it.


> postScannerFilterRow consumes a lot of CPU in tall table scans
> --------------------------------------------------------------
>
>                 Key: HBASE-10047
>                 URL: https://issues.apache.org/jira/browse/HBASE-10047
>             Project: HBase
>          Issue Type: Bug
>            Reporter: Lars Hofhansl
>         Attachments: postScannerFilterRow.png
>
>
> Continuing my profiling quest, I find that in scanning tall table (and filtering everything
on the server) a quarter of the time is now spent in the postScannerFilterRow coprocessor
hook.



--
This message was sent by Atlassian JIRA
(v6.1#6144)

Mime
View raw message