You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@ignite.apache.org by "Atri Sharma (JIRA)" <ji...@apache.org> on 2015/07/06 10:38:04 UTC

[jira] [Created] (IGNITE-1096) Write behind store behaviour in case of abrupt node stop

Atri Sharma created IGNITE-1096:
-----------------------------------

             Summary: Write behind store behaviour in case of abrupt node stop
                 Key: IGNITE-1096
                 URL: https://issues.apache.org/jira/browse/IGNITE-1096
             Project: Ignite
          Issue Type: New Feature
          Components: cache
            Reporter: Atri Sharma


When abrupt stop happens, write behind store should flush data to store. This feature will work with having persistent Write Ahead Logging and have queues on backup nodes which will flush data only in case primary node fails.
This behaviour will be controlled with user defined parameter hence allowing tunable configuration.

Need to minimize performance hit around this.  Structure of WAL record needs to be thought about



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