You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Jonathan Ellis (JIRA)" <ji...@apache.org> on 2010/01/05 00:06:54 UTC
[jira] Created: (CASSANDRA-667) Avoid causing IO contention when
cleaning up compacted SSTables
Avoid causing IO contention when cleaning up compacted SSTables
---------------------------------------------------------------
Key: CASSANDRA-667
URL: https://issues.apache.org/jira/browse/CASSANDRA-667
Project: Cassandra
Issue Type: Improvement
Components: Core
Reporter: Jonathan Ellis
Assignee: Jonathan Ellis
Priority: Minor
Fix For: 0.9
Post-CASSANDRA-408 compacted sstables are deleted in parallel, one thread per sstable; this can cause IO contention with sstable flush and compaction. Serialize this instead.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (CASSANDRA-667) Avoid causing IO contention when
cleaning up compacted SSTables
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-667?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jonathan Ellis updated CASSANDRA-667:
-------------------------------------
Attachment: 667.txt
moved deletions into a (single-threaded) Timer.
> Avoid causing IO contention when cleaning up compacted SSTables
> ---------------------------------------------------------------
>
> Key: CASSANDRA-667
> URL: https://issues.apache.org/jira/browse/CASSANDRA-667
> Project: Cassandra
> Issue Type: Improvement
> Components: Core
> Reporter: Jonathan Ellis
> Assignee: Jonathan Ellis
> Priority: Minor
> Fix For: 0.6
>
> Attachments: 667.txt
>
>
> Post-CASSANDRA-408 compacted sstables are deleted in parallel, one thread per sstable; this can cause IO contention with sstable flush and compaction. Serialize this instead.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-667) Avoid causing IO contention when
cleaning up compacted SSTables
Posted by "Gary Dusbabek (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-667?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12803924#action_12803924 ]
Gary Dusbabek commented on CASSANDRA-667:
-----------------------------------------
+1
> Avoid causing IO contention when cleaning up compacted SSTables
> ---------------------------------------------------------------
>
> Key: CASSANDRA-667
> URL: https://issues.apache.org/jira/browse/CASSANDRA-667
> Project: Cassandra
> Issue Type: Improvement
> Components: Core
> Reporter: Jonathan Ellis
> Assignee: Jonathan Ellis
> Priority: Minor
> Fix For: 0.6
>
> Attachments: 667-v2.txt, 667.txt
>
>
> Post-CASSANDRA-408 compacted sstables are deleted in parallel, one thread per sstable; this can cause IO contention with sstable flush and compaction. Serialize this instead.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Resolved: (CASSANDRA-667) Avoid causing IO contention when
cleaning up compacted SSTables
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-667?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jonathan Ellis resolved CASSANDRA-667.
--------------------------------------
Resolution: Fixed
committed
> Avoid causing IO contention when cleaning up compacted SSTables
> ---------------------------------------------------------------
>
> Key: CASSANDRA-667
> URL: https://issues.apache.org/jira/browse/CASSANDRA-667
> Project: Cassandra
> Issue Type: Improvement
> Components: Core
> Reporter: Jonathan Ellis
> Assignee: Jonathan Ellis
> Priority: Minor
> Fix For: 0.6
>
> Attachments: 667-v2.txt, 667.txt
>
>
> Post-CASSANDRA-408 compacted sstables are deleted in parallel, one thread per sstable; this can cause IO contention with sstable flush and compaction. Serialize this instead.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (CASSANDRA-667) Avoid causing IO contention when
cleaning up compacted SSTables
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-667?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jonathan Ellis updated CASSANDRA-667:
-------------------------------------
Attachment: 667-v2.txt
you're right. updated patch attached.
> Avoid causing IO contention when cleaning up compacted SSTables
> ---------------------------------------------------------------
>
> Key: CASSANDRA-667
> URL: https://issues.apache.org/jira/browse/CASSANDRA-667
> Project: Cassandra
> Issue Type: Improvement
> Components: Core
> Reporter: Jonathan Ellis
> Assignee: Jonathan Ellis
> Priority: Minor
> Fix For: 0.6
>
> Attachments: 667-v2.txt, 667.txt
>
>
> Post-CASSANDRA-408 compacted sstables are deleted in parallel, one thread per sstable; this can cause IO contention with sstable flush and compaction. Serialize this instead.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Assigned: (CASSANDRA-667) Avoid causing IO contention when
cleaning up compacted SSTables
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-667?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jonathan Ellis reassigned CASSANDRA-667:
----------------------------------------
Assignee: Jonathan Ellis (was: Stu Hood)
> Avoid causing IO contention when cleaning up compacted SSTables
> ---------------------------------------------------------------
>
> Key: CASSANDRA-667
> URL: https://issues.apache.org/jira/browse/CASSANDRA-667
> Project: Cassandra
> Issue Type: Improvement
> Components: Core
> Reporter: Jonathan Ellis
> Assignee: Jonathan Ellis
> Priority: Minor
> Fix For: 0.6
>
>
> Post-CASSANDRA-408 compacted sstables are deleted in parallel, one thread per sstable; this can cause IO contention with sstable flush and compaction. Serialize this instead.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Assigned: (CASSANDRA-667) Avoid causing IO contention when
cleaning up compacted SSTables
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-667?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jonathan Ellis reassigned CASSANDRA-667:
----------------------------------------
Assignee: Stu Hood (was: Jonathan Ellis)
> Avoid causing IO contention when cleaning up compacted SSTables
> ---------------------------------------------------------------
>
> Key: CASSANDRA-667
> URL: https://issues.apache.org/jira/browse/CASSANDRA-667
> Project: Cassandra
> Issue Type: Improvement
> Components: Core
> Reporter: Jonathan Ellis
> Assignee: Stu Hood
> Priority: Minor
> Fix For: 0.6
>
>
> Post-CASSANDRA-408 compacted sstables are deleted in parallel, one thread per sstable; this can cause IO contention with sstable flush and compaction. Serialize this instead.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-667) Avoid causing IO contention when
cleaning up compacted SSTables
Posted by "Hudson (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-667?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12804077#action_12804077 ]
Hudson commented on CASSANDRA-667:
----------------------------------
Integrated in Cassandra #332 (See [http://hudson.zones.apache.org/hudson/job/Cassandra/332/])
move sstable deletions into a Timer to reduce concurrency
patch by jbellis; reviewed by gdusbabek for
> Avoid causing IO contention when cleaning up compacted SSTables
> ---------------------------------------------------------------
>
> Key: CASSANDRA-667
> URL: https://issues.apache.org/jira/browse/CASSANDRA-667
> Project: Cassandra
> Issue Type: Improvement
> Components: Core
> Reporter: Jonathan Ellis
> Assignee: Jonathan Ellis
> Priority: Minor
> Fix For: 0.6
>
> Attachments: 667-v2.txt, 667.txt
>
>
> Post-CASSANDRA-408 compacted sstables are deleted in parallel, one thread per sstable; this can cause IO contention with sstable flush and compaction. Serialize this instead.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-667) Avoid causing IO contention when
cleaning up compacted SSTables
Posted by "Gary Dusbabek (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-667?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12803909#action_12803909 ]
Gary Dusbabek commented on CASSANDRA-667:
-----------------------------------------
I don't see the point of passing attempts as a param to CleanupTask. Even if it were ever non-zero, the number wouldn't reflect the number of times it would actually be attempted.
> Avoid causing IO contention when cleaning up compacted SSTables
> ---------------------------------------------------------------
>
> Key: CASSANDRA-667
> URL: https://issues.apache.org/jira/browse/CASSANDRA-667
> Project: Cassandra
> Issue Type: Improvement
> Components: Core
> Reporter: Jonathan Ellis
> Assignee: Jonathan Ellis
> Priority: Minor
> Fix For: 0.6
>
> Attachments: 667.txt
>
>
> Post-CASSANDRA-408 compacted sstables are deleted in parallel, one thread per sstable; this can cause IO contention with sstable flush and compaction. Serialize this instead.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.