[
https://issues.apache.org/jira/browse/TAJO-181?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13767420#comment-13767420
]
Hudson commented on TAJO-181:
-----------------------------
SUCCESS: Integrated in Tajo-trunk-postcommit #432 (See
[https://builds.apache.org/job/Tajo-trunk-postcommit/432/])
TAJO-181: Improvement of disk scheduler policy for straggler disk.
(hyoungjunkim via hyunsik) (hyunsik:
https://git-wip-us.apache.org/repos/asf?p=incubator-tajo.git&a=commit&h=3915d85e98f00c629e064af26dd3551f29c52437)
*
tajo-core/tajo-core-storage/src/main/java/org/apache/tajo/storage/v2/ScanScheduler.java
> Improvement of disk scheduler policy for straggler disk
> -------------------------------------------------------
>
> Key: TAJO-181
> URL: https://issues.apache.org/jira/browse/TAJO-181
> Project: Tajo
> Issue Type: Improvement
> Components: storage
> Affects Versions: 0.2-incubating
> Reporter: hyoungjunkim
> Assignee: hyoungjunkim
> Priority: Minor
> Fix For: 0.2-incubating
>
> Attachments: TAJO-181.patch
>
>
> Current new StorageManager assigns to DiskScanScheduler with disk id
> regardless of each DiskScanScheduler's queue size. This policy is the cause
> of poor performance.
> If some DiskScanScheduler is empty, the DiskScanScheduler is assigned a
> priority.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
