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-12994) Improve network utilization for scanning RPC requests by preloading the next set of results on the server
Date Tue, 10 Feb 2015 04:40:34 GMT

    [ https://issues.apache.org/jira/browse/HBASE-12994?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14313553#comment-14313553
] 

Lars Hofhansl commented on HBASE-12994:
---------------------------------------

Thanks man!

> Improve network utilization for scanning RPC requests by preloading the next set of results
on the server
> ---------------------------------------------------------------------------------------------------------
>
>                 Key: HBASE-12994
>                 URL: https://issues.apache.org/jira/browse/HBASE-12994
>             Project: HBase
>          Issue Type: Improvement
>            Reporter: Jonathan Lawlor
>
> As [~lhofhansl] has called out in HBASE-11544, RPC is inefficient when scanning. The
way it currently works is the client requests a buffer worth of results and works through
that buffer. Once that buffer of results is exhausted, the client realizes it has run out
of results and then requests the next buffer. This could be improved by beginning to load
the next buffer on the server while the client is working through the current buffer of results.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message