You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hbase.apache.org by "Andrew Purtell (JIRA)" <ji...@apache.org> on 2014/02/27 00:47:23 UTC

[jira] [Commented] (HBASE-1734) Priority queue sorted replication policy

    [ https://issues.apache.org/jira/browse/HBASE-1734?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13913755#comment-13913755 ] 

Andrew Purtell commented on HBASE-1734:
---------------------------------------

bq. We have implemented a prototype in a formal work with HBase, and obtained some promising results as discussed in the mailing list. The work can be refined and contributed to the community in the form of a patch if desired. Let me know. 

Yes we would be interested in hearing more. Do you have a publicly available paper, technical report, and/or source code? 

Promoted to top level new feature issue. 

> Priority queue sorted replication policy
> ----------------------------------------
>
>                 Key: HBASE-1734
>                 URL: https://issues.apache.org/jira/browse/HBASE-1734
>             Project: HBase
>          Issue Type: New Feature
>          Components: Replication
>            Reporter: Andrew Purtell
>
> Implement a multi-level priority queue sorted replication policy which plugs into the replication framework (HBASE-1733). Split incoming logs. Sort outbound edits by priority. One option is to consider each column family's replication policy tag as an index into a deadline table for EDF scheduling (http://en.wikipedia.org/wiki/Earliest_deadline_first_scheduling). 



--
This message was sent by Atlassian JIRA
(v6.1.5#6160)