You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@chukwa.apache.org by "Ari Rabkin (JIRA)" <ji...@apache.org> on 2009/09/21 09:58:16 UTC

[jira] Created: (CHUKWA-392) fair queuing for threads in collector

fair queuing for threads in collector
-------------------------------------

                 Key: CHUKWA-392
                 URL: https://issues.apache.org/jira/browse/CHUKWA-392
             Project: Hadoop Chukwa
          Issue Type: Improvement
          Components: data collection
            Reporter: Ari Rabkin


Right now, we don't write data in order received. This means that backpressure doesn't work right.  

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Commented: (CHUKWA-392) fair queuing for threads in collector

Posted by "Hudson (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/CHUKWA-392?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12758297#action_12758297 ] 

Hudson commented on CHUKWA-392:
-------------------------------

Integrated in Chukwa-trunk #145 (See [http://hudson.zones.apache.org/hudson/job/Chukwa-trunk/145/])
    .  FIFO queueing of threads in collector.


> fair queuing for threads in collector
> -------------------------------------
>
>                 Key: CHUKWA-392
>                 URL: https://issues.apache.org/jira/browse/CHUKWA-392
>             Project: Hadoop Chukwa
>          Issue Type: Improvement
>          Components: data collection
>            Reporter: Ari Rabkin
>            Assignee: Ari Rabkin
>         Attachments: CHUKWA-392.patch
>
>
> Right now, we don't write data in order received. This means that backpressure doesn't work right.  

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Commented: (CHUKWA-392) fair queuing for threads in collector

Posted by "Ari Rabkin (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/CHUKWA-392?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12758018#action_12758018 ] 

Ari Rabkin commented on CHUKWA-392:
-----------------------------------

At higher data rates per node, this patch is actually a big win. I'm seeing drastic reduction in retransmits, and a consequent big improvement in goodput.

> fair queuing for threads in collector
> -------------------------------------
>
>                 Key: CHUKWA-392
>                 URL: https://issues.apache.org/jira/browse/CHUKWA-392
>             Project: Hadoop Chukwa
>          Issue Type: Improvement
>          Components: data collection
>            Reporter: Ari Rabkin
>            Assignee: Ari Rabkin
>         Attachments: CHUKWA-392.patch
>
>
> Right now, we don't write data in order received. This means that backpressure doesn't work right.  

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Assigned: (CHUKWA-392) fair queuing for threads in collector

Posted by "Ari Rabkin (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/CHUKWA-392?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Ari Rabkin reassigned CHUKWA-392:
---------------------------------

    Assignee: Ari Rabkin

> fair queuing for threads in collector
> -------------------------------------
>
>                 Key: CHUKWA-392
>                 URL: https://issues.apache.org/jira/browse/CHUKWA-392
>             Project: Hadoop Chukwa
>          Issue Type: Improvement
>          Components: data collection
>            Reporter: Ari Rabkin
>            Assignee: Ari Rabkin
>         Attachments: CHUKWA-392.patch
>
>
> Right now, we don't write data in order received. This means that backpressure doesn't work right.  

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Updated: (CHUKWA-392) fair queuing for threads in collector

Posted by "Ari Rabkin (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/CHUKWA-392?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Ari Rabkin updated CHUKWA-392:
------------------------------

    Status: Patch Available  (was: Open)

Patch ready for review, but I want to test at scale before committing.

> fair queuing for threads in collector
> -------------------------------------
>
>                 Key: CHUKWA-392
>                 URL: https://issues.apache.org/jira/browse/CHUKWA-392
>             Project: Hadoop Chukwa
>          Issue Type: Improvement
>          Components: data collection
>            Reporter: Ari Rabkin
>            Assignee: Ari Rabkin
>         Attachments: CHUKWA-392.patch
>
>
> Right now, we don't write data in order received. This means that backpressure doesn't work right.  

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Commented: (CHUKWA-392) fair queuing for threads in collector

Posted by "Eric Yang (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/CHUKWA-392?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12758019#action_12758019 ] 

Eric Yang commented on CHUKWA-392:
----------------------------------

+1 looks good.

> fair queuing for threads in collector
> -------------------------------------
>
>                 Key: CHUKWA-392
>                 URL: https://issues.apache.org/jira/browse/CHUKWA-392
>             Project: Hadoop Chukwa
>          Issue Type: Improvement
>          Components: data collection
>            Reporter: Ari Rabkin
>            Assignee: Ari Rabkin
>         Attachments: CHUKWA-392.patch
>
>
> Right now, we don't write data in order received. This means that backpressure doesn't work right.  

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Updated: (CHUKWA-392) fair queuing for threads in collector

Posted by "Ari Rabkin (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/CHUKWA-392?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Ari Rabkin updated CHUKWA-392:
------------------------------

    Resolution: Fixed
        Status: Resolved  (was: Patch Available)

I just committed this.

> fair queuing for threads in collector
> -------------------------------------
>
>                 Key: CHUKWA-392
>                 URL: https://issues.apache.org/jira/browse/CHUKWA-392
>             Project: Hadoop Chukwa
>          Issue Type: Improvement
>          Components: data collection
>            Reporter: Ari Rabkin
>            Assignee: Ari Rabkin
>         Attachments: CHUKWA-392.patch
>
>
> Right now, we don't write data in order received. This means that backpressure doesn't work right.  

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Updated: (CHUKWA-392) fair queuing for threads in collector

Posted by "Ari Rabkin (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/CHUKWA-392?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Ari Rabkin updated CHUKWA-392:
------------------------------

    Attachment: CHUKWA-392.patch

> fair queuing for threads in collector
> -------------------------------------
>
>                 Key: CHUKWA-392
>                 URL: https://issues.apache.org/jira/browse/CHUKWA-392
>             Project: Hadoop Chukwa
>          Issue Type: Improvement
>          Components: data collection
>            Reporter: Ari Rabkin
>         Attachments: CHUKWA-392.patch
>
>
> Right now, we don't write data in order received. This means that backpressure doesn't work right.  

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Commented: (CHUKWA-392) fair queuing for threads in collector

Posted by "Ari Rabkin (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/CHUKWA-392?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12757980#action_12757980 ] 

Ari Rabkin commented on CHUKWA-392:
-----------------------------------

OK.  Tested and works ok on 200 nodes each at 100 kb/sec.  Performance indistinguishable from baseline. I propose to commit this.

> fair queuing for threads in collector
> -------------------------------------
>
>                 Key: CHUKWA-392
>                 URL: https://issues.apache.org/jira/browse/CHUKWA-392
>             Project: Hadoop Chukwa
>          Issue Type: Improvement
>          Components: data collection
>            Reporter: Ari Rabkin
>            Assignee: Ari Rabkin
>         Attachments: CHUKWA-392.patch
>
>
> Right now, we don't write data in order received. This means that backpressure doesn't work right.  

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.