[jira] Updated: (HBASE-1050) Allow regions to split around scanners

["Andrew Purtell (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/HBASE-1050?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Andrew Purtell updated HBASE-1050:
----------------------------------

         Priority: Major  (was: Blocker)
    Fix Version/s:     (was: 0.19.0)
                   0.20.0

A configuration change to dfs.datanode.handler.count from its default of 3 to something much larger has altered overall system behavior on my cluster sufficiently to stop the apparent split suppression by scanning mapreduce jobs that I was seeing. 

The core of this issue -- allowing splits around scanners -- is still an important consideration for scalability IMHO. 

Moving out of 0.19 and marking down to major. 

> Allow regions to split around scanners
> --------------------------------------
>
>                 Key: HBASE-1050
>                 URL: https://issues.apache.org/jira/browse/HBASE-1050
>             Project: Hadoop HBase
>          Issue Type: Improvement
>          Components: client, regionserver
>            Reporter: Andrew Purtell
>            Assignee: stack
>             Fix For: 0.20.0
>
>
> We have a number of scanners iterating over a table that also sees a lot of constant write activity. If the scans are too frequent we will suppress splitting. At a lull then a number of splits happen all at once, occasionally overwhelming DFS and causing file corruption. 
> I wonder how much work it would be to split regions around scanners. Rather than wait for scanner leases to expire, suspend/block the scanner, split the table, and then negotiate with the client to continue. 

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.