You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hbase.apache.org by "Nicolas Spiegelberg (JIRA)" <ji...@apache.org> on 2011/08/11 23:22:27 UTC
[jira] [Created] (HBASE-4194) RegionSplitter: Split on low-loaded
regions first
RegionSplitter: Split on low-loaded regions first
-------------------------------------------------
Key: HBASE-4194
URL: https://issues.apache.org/jira/browse/HBASE-4194
Project: HBase
Issue Type: Improvement
Reporter: Nicolas Spiegelberg
Assignee: Nicolas Spiegelberg
Priority: Trivial
Fix For: 0.92.0
When running RegionSplitter, our app devs noticed that they were getting a lot of NSREs. This is caused by 2 factors:
1. the split itself will cause an NSRE
2. any load balancing will cause one.
The former cannot be helped. We can more tightly control load balancing though. Instead of doing a name-sorted round-robin split across RS in the tier, we could sorted the RS by region count. That way, we only split an RS with 10 regions after there are no more RS with 9 regions. This will prevent the load balancing slop from kicking in and will fix the problem where restarting RegionSplitter always starts splitting at RS #1.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HBASE-4194) RegionSplitter: Split on under-loaded
region servers first
Posted by "stack (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-4194?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
stack updated HBASE-4194:
-------------------------
Resolution: Fixed
Hadoop Flags: [Reviewed]
Status: Resolved (was: Patch Available)
This was committed by Ted a while back. Resolving.
> RegionSplitter: Split on under-loaded region servers first
> ----------------------------------------------------------
>
> Key: HBASE-4194
> URL: https://issues.apache.org/jira/browse/HBASE-4194
> Project: HBase
> Issue Type: Improvement
> Reporter: Nicolas Spiegelberg
> Assignee: Nicolas Spiegelberg
> Priority: Trivial
> Fix For: 0.92.0
>
> Attachments: HBASE-4194.patch
>
>
> When running RegionSplitter, our app devs noticed that they were getting a lot of NSREs. This is caused by 2 factors:
> 1. the split itself will cause an NSRE
> 2. any load balancing will cause one.
> The former cannot be helped. We can more tightly control load balancing though. Instead of doing a name-sorted round-robin split across RS in the tier, we could sort the RS's by region count. That way, we only split an RS with 10 regions after there are no more RS with 9 regions. This will prevent the load balancing slop from kicking in and will fix the problem where restarting RegionSplitter always starts splitting at RS #1.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-4194) RegionSplitter: Split on
under-loaded region servers first
Posted by "Hudson (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-4194?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13084736#comment-13084736 ]
Hudson commented on HBASE-4194:
-------------------------------
Integrated in HBase-TRUNK #2113 (See [https://builds.apache.org/job/HBase-TRUNK/2113/])
HBASE-4194 RegionSplitter: Split on under-loaded region servers first
tedyu :
Files :
* /hbase/trunk/src/main/java/org/apache/hadoop/hbase/util/RegionSplitter.java
* /hbase/trunk/CHANGES.txt
> RegionSplitter: Split on under-loaded region servers first
> ----------------------------------------------------------
>
> Key: HBASE-4194
> URL: https://issues.apache.org/jira/browse/HBASE-4194
> Project: HBase
> Issue Type: Improvement
> Reporter: Nicolas Spiegelberg
> Assignee: Nicolas Spiegelberg
> Priority: Trivial
> Fix For: 0.92.0
>
> Attachments: HBASE-4194.patch
>
>
> When running RegionSplitter, our app devs noticed that they were getting a lot of NSREs. This is caused by 2 factors:
> 1. the split itself will cause an NSRE
> 2. any load balancing will cause one.
> The former cannot be helped. We can more tightly control load balancing though. Instead of doing a name-sorted round-robin split across RS in the tier, we could sort the RS's by region count. That way, we only split an RS with 10 regions after there are no more RS with 9 regions. This will prevent the load balancing slop from kicking in and will fix the problem where restarting RegionSplitter always starts splitting at RS #1.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HBASE-4194) RegionSplitter: Split on low-loaded
regions first
Posted by "Nicolas Spiegelberg (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-4194?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Nicolas Spiegelberg updated HBASE-4194:
---------------------------------------
Attachment: HBASE-4194.patch
> RegionSplitter: Split on low-loaded regions first
> -------------------------------------------------
>
> Key: HBASE-4194
> URL: https://issues.apache.org/jira/browse/HBASE-4194
> Project: HBase
> Issue Type: Improvement
> Reporter: Nicolas Spiegelberg
> Assignee: Nicolas Spiegelberg
> Priority: Trivial
> Fix For: 0.92.0
>
> Attachments: HBASE-4194.patch
>
>
> When running RegionSplitter, our app devs noticed that they were getting a lot of NSREs. This is caused by 2 factors:
> 1. the split itself will cause an NSRE
> 2. any load balancing will cause one.
> The former cannot be helped. We can more tightly control load balancing though. Instead of doing a name-sorted round-robin split across RS in the tier, we could sorted the RS by region count. That way, we only split an RS with 10 regions after there are no more RS with 9 regions. This will prevent the load balancing slop from kicking in and will fix the problem where restarting RegionSplitter always starts splitting at RS #1.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-4194) RegionSplitter: Split on low-loaded
regions first
Posted by "Ted Yu (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-4194?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13083879#comment-13083879 ]
Ted Yu commented on HBASE-4194:
-------------------------------
+1 on the patch.
I wonder if we can disable load balancer when RegionSplitter runs.
> RegionSplitter: Split on low-loaded regions first
> -------------------------------------------------
>
> Key: HBASE-4194
> URL: https://issues.apache.org/jira/browse/HBASE-4194
> Project: HBase
> Issue Type: Improvement
> Reporter: Nicolas Spiegelberg
> Assignee: Nicolas Spiegelberg
> Priority: Trivial
> Fix For: 0.92.0
>
> Attachments: HBASE-4194.patch
>
>
> When running RegionSplitter, our app devs noticed that they were getting a lot of NSREs. This is caused by 2 factors:
> 1. the split itself will cause an NSRE
> 2. any load balancing will cause one.
> The former cannot be helped. We can more tightly control load balancing though. Instead of doing a name-sorted round-robin split across RS in the tier, we could sorted the RS by region count. That way, we only split an RS with 10 regions after there are no more RS with 9 regions. This will prevent the load balancing slop from kicking in and will fix the problem where restarting RegionSplitter always starts splitting at RS #1.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HBASE-4194) RegionSplitter: Split on low-loaded
regions first
Posted by "Nicolas Spiegelberg (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-4194?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Nicolas Spiegelberg updated HBASE-4194:
---------------------------------------
Status: Patch Available (was: Open)
> RegionSplitter: Split on low-loaded regions first
> -------------------------------------------------
>
> Key: HBASE-4194
> URL: https://issues.apache.org/jira/browse/HBASE-4194
> Project: HBase
> Issue Type: Improvement
> Reporter: Nicolas Spiegelberg
> Assignee: Nicolas Spiegelberg
> Priority: Trivial
> Fix For: 0.92.0
>
> Attachments: HBASE-4194.patch
>
>
> When running RegionSplitter, our app devs noticed that they were getting a lot of NSREs. This is caused by 2 factors:
> 1. the split itself will cause an NSRE
> 2. any load balancing will cause one.
> The former cannot be helped. We can more tightly control load balancing though. Instead of doing a name-sorted round-robin split across RS in the tier, we could sorted the RS by region count. That way, we only split an RS with 10 regions after there are no more RS with 9 regions. This will prevent the load balancing slop from kicking in and will fix the problem where restarting RegionSplitter always starts splitting at RS #1.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HBASE-4194) RegionSplitter: Split on under-loaded
region servers first
Posted by "Ted Yu (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-4194?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Ted Yu updated HBASE-4194:
--------------------------
Description:
When running RegionSplitter, our app devs noticed that they were getting a lot of NSREs. This is caused by 2 factors:
1. the split itself will cause an NSRE
2. any load balancing will cause one.
The former cannot be helped. We can more tightly control load balancing though. Instead of doing a name-sorted round-robin split across RS in the tier, we could sort the RS's by region count. That way, we only split an RS with 10 regions after there are no more RS with 9 regions. This will prevent the load balancing slop from kicking in and will fix the problem where restarting RegionSplitter always starts splitting at RS #1.
was:
When running RegionSplitter, our app devs noticed that they were getting a lot of NSREs. This is caused by 2 factors:
1. the split itself will cause an NSRE
2. any load balancing will cause one.
The former cannot be helped. We can more tightly control load balancing though. Instead of doing a name-sorted round-robin split across RS in the tier, we could sorted the RS by region count. That way, we only split an RS with 10 regions after there are no more RS with 9 regions. This will prevent the load balancing slop from kicking in and will fix the problem where restarting RegionSplitter always starts splitting at RS #1.
Summary: RegionSplitter: Split on under-loaded region servers first (was: RegionSplitter: Split on low-loaded regions first)
> RegionSplitter: Split on under-loaded region servers first
> ----------------------------------------------------------
>
> Key: HBASE-4194
> URL: https://issues.apache.org/jira/browse/HBASE-4194
> Project: HBase
> Issue Type: Improvement
> Reporter: Nicolas Spiegelberg
> Assignee: Nicolas Spiegelberg
> Priority: Trivial
> Fix For: 0.92.0
>
> Attachments: HBASE-4194.patch
>
>
> When running RegionSplitter, our app devs noticed that they were getting a lot of NSREs. This is caused by 2 factors:
> 1. the split itself will cause an NSRE
> 2. any load balancing will cause one.
> The former cannot be helped. We can more tightly control load balancing though. Instead of doing a name-sorted round-robin split across RS in the tier, we could sort the RS's by region count. That way, we only split an RS with 10 regions after there are no more RS with 9 regions. This will prevent the load balancing slop from kicking in and will fix the problem where restarting RegionSplitter always starts splitting at RS #1.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-4194) RegionSplitter: Split on
under-loaded region servers first
Posted by "Ted Yu (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-4194?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13084607#comment-13084607 ]
Ted Yu commented on HBASE-4194:
-------------------------------
Integrated to TRUNK.
Thanks for the patch Nicholas.
> RegionSplitter: Split on under-loaded region servers first
> ----------------------------------------------------------
>
> Key: HBASE-4194
> URL: https://issues.apache.org/jira/browse/HBASE-4194
> Project: HBase
> Issue Type: Improvement
> Reporter: Nicolas Spiegelberg
> Assignee: Nicolas Spiegelberg
> Priority: Trivial
> Fix For: 0.92.0
>
> Attachments: HBASE-4194.patch
>
>
> When running RegionSplitter, our app devs noticed that they were getting a lot of NSREs. This is caused by 2 factors:
> 1. the split itself will cause an NSRE
> 2. any load balancing will cause one.
> The former cannot be helped. We can more tightly control load balancing though. Instead of doing a name-sorted round-robin split across RS in the tier, we could sort the RS's by region count. That way, we only split an RS with 10 regions after there are no more RS with 9 regions. This will prevent the load balancing slop from kicking in and will fix the problem where restarting RegionSplitter always starts splitting at RS #1.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira