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-9753) Excessive readpoints checks in MemstoreScanner
Date Sun, 13 Oct 2013 01:58:42 GMT

    [ https://issues.apache.org/jira/browse/HBASE-9753?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13793559#comment-13793559
] 

Lars Hofhansl commented on HBASE-9753:
--------------------------------------

I think I was too eager filing tickets :)

Yep, we should be able to completely do away with MultiVersionConsistencyControl.perThreadReadPoint.

In the same way you suggested to pass the readpoint to the MemstoreScanner consructor we should
be able to pass the same to all StoreFileScanners. That would not only be more performant,
it would also be easier on in the eye.

If you have time to do a patch here or on HBASE-9754 that'd be awesome, otherwise I'll pick
it up next week.


> Excessive readpoints checks in MemstoreScanner
> ----------------------------------------------
>
>                 Key: HBASE-9753
>                 URL: https://issues.apache.org/jira/browse/HBASE-9753
>             Project: HBase
>          Issue Type: Bug
>            Reporter: Lars Hofhansl
>            Assignee: Vladimir Rodionov
>
> Brought up by [~vrodionov] on the mailing list. See also HBASE-9751.



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

Mime
View raw message