You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@phoenix.apache.org by "James Taylor (JIRA)" <ji...@apache.org> on 2014/11/15 03:35:33 UTC
[jira] [Commented] (PHOENIX-1455) Replace org.xerial.snappy with
org.iq80.snappy pure Java snappy implementation
[ https://issues.apache.org/jira/browse/PHOENIX-1455?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14213292#comment-14213292 ]
James Taylor commented on PHOENIX-1455:
---------------------------------------
This is awesome! Great find, @Andrew. It seems we don't have a test that exercising this compression logic (correct me if I'm wrong, [~anoopsamjohn] or [~maryannxue]). We should set the QueryServicesOptions.DISTINCT_VALUE_COMPRESS_THRESHOLD_ATTRIB (threshold for when compression occurs) to a low value for one of the DistinctCountIT and HashJoinIT tests (or have a derived/new test class that sets this config parameter so the tests get exercised with compression).
Any thoughts on adding those tests before/after/with this change?
> Replace org.xerial.snappy with org.iq80.snappy pure Java snappy implementation
> ------------------------------------------------------------------------------
>
> Key: PHOENIX-1455
> URL: https://issues.apache.org/jira/browse/PHOENIX-1455
> Project: Phoenix
> Issue Type: Improvement
> Reporter: Andrew Purtell
> Assignee: Andrew Purtell
> Fix For: 5.0.0, 4.2, 3.2
>
> Attachments: PHOENIX-1455.patch
>
>
> We can avoid Phoenix runtime dependencies on native snappy support by using the all Java snappy compression implementation found at https://github.com/dain/snappy.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)