You are viewing a plain text version of this content. The canonical link for it is here.
Posted to common-dev@hadoop.apache.org by "Hairong Kuang (JIRA)" <ji...@apache.org> on 2008/08/13 02:07:44 UTC
[jira] Created: (HADOOP-3943) NetworkTopology.pseudoSortByDistance
does not need to be a synchronized method
NetworkTopology.pseudoSortByDistance does not need to be a synchronized method
------------------------------------------------------------------------------
Key: HADOOP-3943
URL: https://issues.apache.org/jira/browse/HADOOP-3943
Project: Hadoop Core
Issue Type: Improvement
Components: dfs
Reporter: Hairong Kuang
Assignee: Hairong Kuang
Fix For: 0.19.0
When all requests to namenode are read, most handlers block on NetworkTopology object lock when sorting replicas. Removing this necessary lock would reduce the contention in namenode and therefore improve its throughput.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-3943)
NetworkTopology.pseudoSortByDistance does not need to be a synchronized
method
Posted by "dhruba borthakur (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3943?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12628059#action_12628059 ]
dhruba borthakur commented on HADOOP-3943:
------------------------------------------
+1. Looks perfect to me.
> NetworkTopology.pseudoSortByDistance does not need to be a synchronized method
> ------------------------------------------------------------------------------
>
> Key: HADOOP-3943
> URL: https://issues.apache.org/jira/browse/HADOOP-3943
> Project: Hadoop Core
> Issue Type: Improvement
> Components: dfs
> Reporter: Hairong Kuang
> Assignee: Hairong Kuang
> Fix For: 0.19.0
>
> Attachments: sortDistance.patch
>
>
> When all requests to namenode are read, most handlers block on NetworkTopology object lock when sorting replicas. Removing this necessary lock would reduce the contention in namenode and therefore improve its throughput.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-3943) NetworkTopology.pseudoSortByDistance
does not need to be a synchronized method
Posted by "Hairong Kuang (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3943?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Hairong Kuang updated HADOOP-3943:
----------------------------------
Hadoop Flags: [Reviewed]
Status: Patch Available (was: Open)
> NetworkTopology.pseudoSortByDistance does not need to be a synchronized method
> ------------------------------------------------------------------------------
>
> Key: HADOOP-3943
> URL: https://issues.apache.org/jira/browse/HADOOP-3943
> Project: Hadoop Core
> Issue Type: Improvement
> Components: dfs
> Reporter: Hairong Kuang
> Assignee: Hairong Kuang
> Fix For: 0.19.0
>
> Attachments: sortDistance.patch
>
>
> When all requests to namenode are read, most handlers block on NetworkTopology object lock when sorting replicas. Removing this necessary lock would reduce the contention in namenode and therefore improve its throughput.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-3943) NetworkTopology.pseudoSortByDistance
does not need to be a synchronized method
Posted by "Hairong Kuang (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3943?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Hairong Kuang updated HADOOP-3943:
----------------------------------
Attachment: sortDistance.patch
It seems to me that simply removing "synchronized" should work. There is no need to grab the read lock.
> NetworkTopology.pseudoSortByDistance does not need to be a synchronized method
> ------------------------------------------------------------------------------
>
> Key: HADOOP-3943
> URL: https://issues.apache.org/jira/browse/HADOOP-3943
> Project: Hadoop Core
> Issue Type: Improvement
> Components: dfs
> Reporter: Hairong Kuang
> Assignee: Hairong Kuang
> Fix For: 0.19.0
>
> Attachments: sortDistance.patch
>
>
> When all requests to namenode are read, most handlers block on NetworkTopology object lock when sorting replicas. Removing this necessary lock would reduce the contention in namenode and therefore improve its throughput.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-3943) NetworkTopology.pseudoSortByDistance
does not need to be a synchronized method
Posted by "Owen O'Malley (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3943?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Owen O'Malley updated HADOOP-3943:
----------------------------------
Resolution: Fixed
Status: Resolved (was: Patch Available)
I just committed this. Thanks, Hairong!
> NetworkTopology.pseudoSortByDistance does not need to be a synchronized method
> ------------------------------------------------------------------------------
>
> Key: HADOOP-3943
> URL: https://issues.apache.org/jira/browse/HADOOP-3943
> Project: Hadoop Core
> Issue Type: Improvement
> Components: dfs
> Reporter: Hairong Kuang
> Assignee: Hairong Kuang
> Fix For: 0.19.0
>
> Attachments: sortDistance.patch
>
>
> When all requests to namenode are read, most handlers block on NetworkTopology object lock when sorting replicas. Removing this necessary lock would reduce the contention in namenode and therefore improve its throughput.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-3943)
NetworkTopology.pseudoSortByDistance does not need to be a synchronized
method
Posted by "Raghu Angadi (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3943?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12628064#action_12628064 ]
Raghu Angadi commented on HADOOP-3943:
--------------------------------------
That keeps it "as correct as before", which is fine. +1
> NetworkTopology.pseudoSortByDistance does not need to be a synchronized method
> ------------------------------------------------------------------------------
>
> Key: HADOOP-3943
> URL: https://issues.apache.org/jira/browse/HADOOP-3943
> Project: Hadoop Core
> Issue Type: Improvement
> Components: dfs
> Reporter: Hairong Kuang
> Assignee: Hairong Kuang
> Fix For: 0.19.0
>
> Attachments: sortDistance.patch
>
>
> When all requests to namenode are read, most handlers block on NetworkTopology object lock when sorting replicas. Removing this necessary lock would reduce the contention in namenode and therefore improve its throughput.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-3943)
NetworkTopology.pseudoSortByDistance does not need to be a synchronized
method
Posted by "Raghu Angadi (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3943?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12622368#action_12622368 ]
Raghu Angadi commented on HADOOP-3943:
--------------------------------------
I guess we missed converting this to use read lock as part of HADOOP-1269. Would that be the fix?
> NetworkTopology.pseudoSortByDistance does not need to be a synchronized method
> ------------------------------------------------------------------------------
>
> Key: HADOOP-3943
> URL: https://issues.apache.org/jira/browse/HADOOP-3943
> Project: Hadoop Core
> Issue Type: Improvement
> Components: dfs
> Reporter: Hairong Kuang
> Assignee: Hairong Kuang
> Fix For: 0.19.0
>
>
> When all requests to namenode are read, most handlers block on NetworkTopology object lock when sorting replicas. Removing this necessary lock would reduce the contention in namenode and therefore improve its throughput.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-3943)
NetworkTopology.pseudoSortByDistance does not need to be a synchronized
method
Posted by "Hudson (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3943?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12628864#action_12628864 ]
Hudson commented on HADOOP-3943:
--------------------------------
Integrated in Hadoop-trunk #595 (See [http://hudson.zones.apache.org/hudson/job/Hadoop-trunk/595/])
> NetworkTopology.pseudoSortByDistance does not need to be a synchronized method
> ------------------------------------------------------------------------------
>
> Key: HADOOP-3943
> URL: https://issues.apache.org/jira/browse/HADOOP-3943
> Project: Hadoop Core
> Issue Type: Improvement
> Components: dfs
> Reporter: Hairong Kuang
> Assignee: Hairong Kuang
> Fix For: 0.19.0
>
> Attachments: sortDistance.patch
>
>
> When all requests to namenode are read, most handlers block on NetworkTopology object lock when sorting replicas. Removing this necessary lock would reduce the contention in namenode and therefore improve its throughput.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.