You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hbase.apache.org by "stack (JIRA)" <ji...@apache.org> on 2011/04/20 21:05:05 UTC
[jira] [Created] (HBASE-3803) Make the load balancer run with a
gentle hand
Make the load balancer run with a gentle hand
---------------------------------------------
Key: HBASE-3803
URL: https://issues.apache.org/jira/browse/HBASE-3803
Project: HBase
Issue Type: Improvement
Reporter: stack
We need 'smoothing' of balancer region move.... Yesterday we brought a regionserver back online into a smallish cluster that was under load and the balance run unloaded a bunch of regions all in the one go which put a dent in the throughput when a bunch of regions went offline at the one time. It'd be sweet if the balancer ran at a context appropriate 'rate'; when under load, it should move regions 'gently' rather than all as a big bang (the decommission script will move a region at a time, verifying it deployed in its new location before moving another... this can take ages to complete but its proven minimally disruptive to loadings)
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-3803) Make the load balancer run with a
gentle hand
Posted by "stack (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-3803?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13450297#comment-13450297 ]
stack commented on HBASE-3803:
------------------------------
HBASE-6720 seems to be a much less that his issue asks for (context appropriate rate)
> Make the load balancer run with a gentle hand
> ---------------------------------------------
>
> Key: HBASE-3803
> URL: https://issues.apache.org/jira/browse/HBASE-3803
> Project: HBase
> Issue Type: Improvement
> Reporter: stack
>
> We need 'smoothing' of balancer region move.... Yesterday we brought a regionserver back online into a smallish cluster that was under load and the balance run unloaded a bunch of regions all in the one go which put a dent in the throughput when a bunch of regions went offline at the one time. It'd be sweet if the balancer ran at a context appropriate 'rate'; when under load, it should move regions 'gently' rather than all as a big bang (the decommission script will move a region at a time, verifying it deployed in its new location before moving another... this can take ages to complete but its proven minimally disruptive to loadings)
--
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
[jira] [Commented] (HBASE-3803) Make the load balancer run with a
gentle hand
Posted by "Lars Hofhansl (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-3803?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13449085#comment-13449085 ]
Lars Hofhansl commented on HBASE-3803:
--------------------------------------
Can we subsume this in HBASE-6720.
> Make the load balancer run with a gentle hand
> ---------------------------------------------
>
> Key: HBASE-3803
> URL: https://issues.apache.org/jira/browse/HBASE-3803
> Project: HBase
> Issue Type: Improvement
> Reporter: stack
>
> We need 'smoothing' of balancer region move.... Yesterday we brought a regionserver back online into a smallish cluster that was under load and the balance run unloaded a bunch of regions all in the one go which put a dent in the throughput when a bunch of regions went offline at the one time. It'd be sweet if the balancer ran at a context appropriate 'rate'; when under load, it should move regions 'gently' rather than all as a big bang (the decommission script will move a region at a time, verifying it deployed in its new location before moving another... this can take ages to complete but its proven minimally disruptive to loadings)
--
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