You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@hbase.apache.org by "stack (JIRA)" <ji...@apache.org> on 2016/01/25 19:24:39 UTC

[jira] [Resolved] (HBASE-7408) Make compaction to use pread instead of sequential read

     [ https://issues.apache.org/jira/browse/HBASE-7408?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

stack resolved HBASE-7408.
--------------------------
    Resolution: Unresolved

Resolving as 'unresolved'.

What is wrong here? Complaint is that sequential reads caused unwarranted i/o but don't we want the HDFS pipelining noted here HBASE-5979 when doing compactions?

> Make compaction to use pread instead of sequential read
> -------------------------------------------------------
>
>                 Key: HBASE-7408
>                 URL: https://issues.apache.org/jira/browse/HBASE-7408
>             Project: HBase
>          Issue Type: Improvement
>            Reporter: Rishit Shroff
>            Assignee: Rishit Shroff
>            Priority: Minor
>   Original Estimate: 24h
>  Remaining Estimate: 24h
>
> As we discovered lately, HFile compactions use sequential reads to fetch blocks. It cause unwanted streaming of data from HDFS to region servers when there is a cache hit. Lets change to use preads to reduce iops on disks.



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