[
https://issues.apache.org/jira/browse/GORA-89?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13545492#comment-13545492
]
Alfonso Nishikawa commented on GORA-89:
---------------------------------------
Hi, Henry.
I checked the patch in a [email protected] with 4GB and took a bit more than 33 minutes
to test gora-hbase.
What I noticed is that creating/enable and disabling tables takes much time. Is
it possible to avoid creating/disabling tables between tests?
If number of rows is low (I think they are), I suggest delete each row (gets
marked to delete) instead of disable+drop (noticed in [0]).
[0] - http://stackoverflow.com/questions/9902353/how-to-clear-a-table-in-hbase
> Avoid HBase MiniCluster restarts to shorten gora-hbase tests
> ------------------------------------------------------------
>
> Key: GORA-89
> URL: https://issues.apache.org/jira/browse/GORA-89
> Project: Apache Gora
> Issue Type: Improvement
> Components: storage-hbase
> Affects Versions: 0.2
> Reporter: Lewis John McGibbney
> Priority: Critical
> Fix For: 0.3
>
> Attachments: GORA-89-hsaputra.patch, GORA-89.patch
>
>
> Currently our hbase tests are taking forever and a day. We should shorten the
> time by avoiding MiniCluster restarts.
> Just implement the cluster as a singleton and clean up the tables in
> between test by doing a scan and deletes for all rows. It's much
> faster than restarting the cluster.
> For code referenece please see the implementation here[1]. The class is
> HBaseClusterSingleton. It needs some refactoring but I think it's
> enough to speed your tests.
> Thanks Ioan for the heads up.
> [1]
> http://svn.apache.org/repos/asf/james/mailbox/trunk/hbase/src/test/java/org/apache/james/mailbox/hbase/
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
