You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hbase.apache.org by "Lars Hofhansl (JIRA)" <ji...@apache.org> on 2018/04/20 02:35:00 UTC

[jira] [Comment Edited] (HBASE-20459) Majority of scan time in HBase-1 spent in size estimation

    [ https://issues.apache.org/jira/browse/HBASE-20459?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16445145#comment-16445145 ] 

Lars Hofhansl edited comment on HBASE-20459 at 4/20/18 2:34 AM:
----------------------------------------------------------------

If the region server is not busy the query time will actually not increase, but I see that the region server consumes between 25% and 40% more CPU time.

So it's quite easy to make the region server CPU bound with just a few concurrent scans.


was (Author: lhofhansl):
If the region server is not busy the query time will actually not increase, but I see that the region server consumes between 25% and 40% more CPU time.

> Majority of scan time in HBase-1 spent in size estimation
> ---------------------------------------------------------
>
>                 Key: HBASE-20459
>                 URL: https://issues.apache.org/jira/browse/HBASE-20459
>             Project: HBase
>          Issue Type: Improvement
>    Affects Versions: 1.4.3
>            Reporter: Lars Hofhansl
>            Priority: Major
>             Fix For: 1.5.0, 1.4.4
>
>         Attachments: Screenshot_20180419_162559.png
>
>
> See attached screenshot. Will look into a fix later.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)