[ 
https://issues.apache.org/jira/browse/HBASE-13965?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Lei Chen updated HBASE-13965:
-----------------------------
    Attachment: HBASE-13965-v7.patch

Updates:
1. Overloaded balanceCluster() to pass the table name to balancer
2. Moved some string constants to HConstants.java
3. Stochastic balancer auto adjust JMX metrics size by the number of tables
4. Stochastic handles both ensemble and per-table modes.
5. Updated tests to cover both modes.

TODO:
1. The tests currently only use the miniCluster to save and read JMX metrics, 
which means that the tables are not actually stored in hbase. I'm NOT sure if 
this method is adequate or we need to save real tables to miniCluster and 
balance them for real?

Sorry guys, I still cannot upload the patch file to review board. The diff file 
always gets "No valid separator after the filename was found in the diff 
header" error. If I manually touch up the file by adding (revision xxxx) or 
(working copy), I got "revision xxxx cannot be found" error. The command line 
rbt has the same problem.

> Stochastic Load Balancer JMX Metrics
> ------------------------------------
>
>                 Key: HBASE-13965
>                 URL: https://issues.apache.org/jira/browse/HBASE-13965
>             Project: HBase
>          Issue Type: Improvement
>          Components: Balancer, metrics
>            Reporter: Lei Chen
>            Assignee: Lei Chen
>         Attachments: HBASE-13965-v3.patch, HBASE-13965-v4.patch, 
> HBASE-13965-v5.patch, HBASE-13965-v6.patch, HBASE-13965-v7.patch, 
> HBASE-13965_v2.patch, HBase-13965-v1.patch, 
> stochasticloadbalancerclasses_v2.png
>
>
> Today’s default HBase load balancer (the Stochastic load balancer) is cost 
> function based. The cost function weights are tunable but no visibility into 
> those cost function results is directly provided.
> A driving example is a cluster we have been tuning which has skewed rack size 
> (one rack has half the nodes of the other few racks). We are tuning the 
> cluster for uniform response time from all region servers with the ability to 
> tolerate a rack failure. Balancing LocalityCost, RegionReplicaRack Cost and 
> RegionCountSkew Cost is difficult without a way to attribute each cost 
> function’s contribution to overall cost. 
> What this jira proposes is to provide visibility via JMX into each cost 
> function of the stochastic load balancer, as well as the overall cost of the 
> balancing plan.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Reply via email to