[ https://issues.apache.org/jira/browse/HBASE-17462?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15830861#comment-15830861 ]
Tim Brown edited comment on HBASE-17462 at 1/20/17 12:05 AM: ------------------------------------------------------------- The regionLoadList has a configurable size. By default it is 15 and controlled by "hbase.master.balancer.stochastic.numRegionLoadsToRemember". For performance, I can attach some screenshots of charts that look at the average number of requests per region server. If you want more in depth analysis let me know. was (Author: timbrown): The regionLoadList has a configurable size. By default it is 15 and controlled by `hbase.master.balancer.stochastic.numRegionLoadsToRemember`. For performance, I can attach some screenshots of charts that look at the average number of requests per region server. If you want more in depth analysis let me know. > Investigate using sliding window for read/write request costs in > StochasticLoadBalancer > --------------------------------------------------------------------------------------- > > Key: HBASE-17462 > URL: https://issues.apache.org/jira/browse/HBASE-17462 > Project: HBase > Issue Type: Improvement > Reporter: Ted Yu > Assignee: Tim Brown > Labels: patch > Attachments: HBASE-17462.patch > > > In the thread, http://search-hadoop.com/m/HBase/YGbbyUZKXWALkX1, Timothy was > asking whether the read/write request costs in StochasticLoadBalancer should > be calculated as rates. > This makes sense since read / write load on region server tends to fluctuate > over time. Using sliding window would reflect more recent trend in read / > write load. > Some factors to consider: > The data structure used by StochasticLoadBalancer should be concise. The > number of regions in a cluster can be expected to approach 1 million. We > cannot afford to store long history of read / write requests in master. > Efficiency of cost calculation should be high - there're many cost > functions the balancer goes through, it is expected for each cost function > to return quickly. Otherwise we would not come up with proper region > movement plan(s) in time. -- This message was sent by Atlassian JIRA (v6.3.4#6332)