You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hbase.apache.org by "Enis Soztutar (JIRA)" <ji...@apache.org> on 2013/03/15 03:54:12 UTC
[jira] [Commented] (HBASE-8119) StochasticLoadBalancer does not
take into account per table balance
[ https://issues.apache.org/jira/browse/HBASE-8119?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13603042#comment-13603042 ]
Enis Soztutar commented on HBASE-8119:
--------------------------------------
Relevant logs from master:
{code}
2013-03-14 21:48:58,078 DEBUG org.apache.hadoop.hbase.master.balancer.StochasticLoadBalancer: Could not find a better load balance plan. Tried 15000 different configurations in 1173296ms, and did not find anything with a computed cost less than 1.7643979505427219
2013-03-14 22:08:47,935 DEBUG org.apache.hadoop.hbase.master.balancer.StochasticLoadBalancer: Could not find a better load balance plan. Tried 15000 different configurations in 1189807ms, and did not find anything with a computed cost less than 1.7654704688906333
2013-03-14 22:28:47,551 DEBUG org.apache.hadoop.hbase.master.balancer.StochasticLoadBalancer: Could not find a better load balance plan. Tried 15000 different configurations in 1199570ms, and did not find anything with a computed cost less than 1.7654704688906333
2013-03-14 22:47:03,075 DEBUG org.apache.hadoop.hbase.master.balancer.StochasticLoadBalancer: Could not find a better load balance plan. Tried 15000 different configurations in 1095479ms, and did not find anything with a computed cost less than 1.7654704688906333
2013-03-14 23:02:14,412 DEBUG org.apache.hadoop.hbase.master.balancer.StochasticLoadBalancer: Could not find a better load balance plan. Tried 15000 different configurations in 911294ms, and did not find anything with a computed cost less than 1.7654704688906333
2013-03-14 23:17:29,509 DEBUG org.apache.hadoop.hbase.master.balancer.StochasticLoadBalancer: Could not find a better load balance plan. Tried 15000 different configurations in 915081ms, and did not find anything with a computed cost less than 1.7654704688906333
2013-03-14 23:30:52,330 DEBUG org.apache.hadoop.hbase.master.balancer.StochasticLoadBalancer: Finished computing new laod balance plan. Computation took 802783ms to try 15000 different iterations. Found a solution that moves 3 regions; Going from a computed cost of 1.8511422856730866 to a new cost of 1.7029355346182344
2013-03-14 23:42:27,086 DEBUG org.apache.hadoop.hbase.master.balancer.StochasticLoadBalancer: Finished computing new laod balance plan. Computation took 694752ms to try 15000 different iterations. Found a solution that moves 50 regions; Going from a computed cost of 23.32442143294416 to a new cost of 8.247462558485287
2013-03-14 23:54:24,813 DEBUG org.apache.hadoop.hbase.master.balancer.StochasticLoadBalancer: Finished computing new laod balance plan. Computation took 717719ms to try 15000 different iterations. Found a solution that moves 110 regions; Going from a computed cost of 32.33871532513386 to a new cost of 8.900759238202657
2013-03-15 00:11:50,998 DEBUG org.apache.hadoop.hbase.master.balancer.StochasticLoadBalancer: Finished computing new laod balance plan. Computation took 746160ms to try 15000 different iterations. Found a solution that moves 148 regions; Going from a computed cost of 37.00339551208993 to a new cost of 8.989686053539122
2013-03-15 00:23:58,508 DEBUG org.apache.hadoop.hbase.master.balancer.StochasticLoadBalancer: Finished computing new laod balance plan. Computation took 727505ms to try 15000 different iterations. Found a solution that moves 181 regions; Going from a computed cost of 35.14975447740123 to a new cost of 4.061124062359311
2013-03-15 00:41:34,881 DEBUG org.apache.hadoop.hbase.master.balancer.StochasticLoadBalancer: Finished computing new laod balance plan. Computation took 755200ms to try 15000 different iterations. Found a solution that moves 198 regions; Going from a computed cost of 36.14252424853199 to a new cost of 4.094303080576197
2013-03-15 00:59:27,394 DEBUG org.apache.hadoop.hbase.master.balancer.StochasticLoadBalancer: Finished computing new laod balance plan. Computation took 772487ms to try 15000 different iterations. Found a solution that moves 209 regions; Going from a computed cost of 36.57889975126676 to a new cost of 4.210851788219407
2013-03-15 01:17:36,289 DEBUG org.apache.hadoop.hbase.master.balancer.StochasticLoadBalancer: Finished computing new laod balance plan. Computation took 788868ms to try 15000 different iterations. Found a solution that moves 209 regions; Going from a computed cost of 36.56674893820493 to a new cost of 3.9487649893586143
2013-03-15 01:36:12,271 DEBUG org.apache.hadoop.hbase.master.balancer.StochasticLoadBalancer: Finished computing new laod balance plan. Computation took 815147ms to try 15000 different iterations. Found a solution that moves 208 regions; Going from a computed cost of 41.39462651313847 to a new cost of 8.942383869697869
2013-03-15 01:55:07,721 DEBUG org.apache.hadoop.hbase.master.balancer.StochasticLoadBalancer: Finished computing new laod balance plan. Computation took 835281ms to try 15000 different iterations. Found a solution that moves 208 regions; Going from a computed cost of 36.22250344352618 to a new cost of 3.9835014102059603
2013-03-15 02:14:15,914 DEBUG org.apache.hadoop.hbase.master.balancer.StochasticLoadBalancer: Finished computing new laod balance plan. Computation took 846886ms to try 15000 different iterations. Found a solution that moves 208 regions; Going from a computed cost of 40.818599028072306 to a new cost of 9.14190871601471
2013-03-15 02:33:08,593 DEBUG org.apache.hadoop.hbase.master.balancer.StochasticLoadBalancer: Finished computing new laod balance plan. Computation took 831429ms to try 15000 different iterations. Found a solution that moves 208 regions; Going from a computed cost of 35.29205428038984 to a new cost of 3.8843714841308463
{code}
It seems that after a balancer is run, only 1 region moves from the logs.
> StochasticLoadBalancer does not take into account per table balance
> -------------------------------------------------------------------
>
> Key: HBASE-8119
> URL: https://issues.apache.org/jira/browse/HBASE-8119
> Project: HBase
> Issue Type: Bug
> Components: Region Assignment
> Affects Versions: 0.98.0
> Reporter: Enis Soztutar
> Fix For: 0.95.0
>
>
> On a 5 node trunk cluster, I ran into a weird problem with StochasticLoadBalancer:
> server1 Thu Mar 14 03:42:50 UTC 2013 0.0 33
> server2 Thu Mar 14 03:47:53 UTC 2013 0.0 34
> server3 Thu Mar 14 03:46:53 UTC 2013 465.0 42
> server4 Thu Mar 14 03:47:53 UTC 2013 11455.0 282
> server5 Thu Mar 14 03:47:53 UTC 2013 0.0 34
> Total:5 11920 425
> Notice that server4 has 282 regions, while the others have much less. Plus for one table with 260 regions has been super imbalanced:
> {code}
> Regions by Region Server
> Region Server Region Count
> http://server3:60030/ 10
> http://server4:60030/ 250
> {code}
--
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