[
https://issues.apache.org/jira/browse/HBASE-8803?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13724713#comment-13724713
]
Nick Dimiduk commented on HBASE-8803:
-------------------------------------
Here's some run numbers. The summary is that the patch makes a huge difference,
but there's a sweets-spot re: the number of threads. That spot appears to hover
somewhere close to the number of RegionServers in the cluster, as you implied
earlier in the conversation.
At least for 0.94, I say clean up the extra logging and ship it. I haven't
tested vs trunk yet.
{noformat}
# 9 region servers, 3000 regions total:
## 0.94.6.1 + 8803v4, maxthreads 1 (default)
real 425m10.306s
user 102m30.018s
sys 86m17.567s
real 324m21.502s
user 78m41.326s
sys 64m28.686s
real 430m2.716s
user 103m11.999s
sys 87m8.496s
## 0.94.6.1 + 8803v4, maxthreads 10
real 205m36.658s
user 92m56.465s
sys 75m18.526s
real 205m9.664s
user 87m53.164s
sys 70m0.675s
real 208m52.563s
user 96m7.719s
sys 79m25.761s
## 0.94.6.1 + 8803v4, maxthreads 30
real 250m38.159s
user 95m19.163s
sys 77m18.219s
real 205m15.578s
user 97m10.359s
sys 79m23.973s
real 242m33.605s
user 92m17.179s
sys 74m23.087s
{noformat}
> region_mover.rb should move multiple regions at a time
> ------------------------------------------------------
>
> Key: HBASE-8803
> URL: https://issues.apache.org/jira/browse/HBASE-8803
> Project: HBase
> Issue Type: Bug
> Components: Usability
> Affects Versions: 0.98.0, 0.94.8, 0.95.1
> Reporter: Jean-Marc Spaggiari
> Assignee: Jean-Marc Spaggiari
> Priority: Critical
> Fix For: 0.98.0, 0.95.2, 0.94.11
>
> Attachments: HBASE-8803-v0-trunk.patch, HBASE-8803-v1-0.94.patch,
> HBASE-8803-v1-trunk.patch, HBASE-8803-v2-0.94.patch,
> HBASE-8803-v2-0.94.patch, HBASE-8803-v3-0.94.patch, HBASE-8803-v4-0.94.patch
>
> Original Estimate: 48h
> Remaining Estimate: 48h
>
> When there is many regions in a cluster, rolling_restart can take hours
> because region_mover is moving the regions one by one.
--
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
