drill-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Brian ONeill (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (DRILL-16) Cassandra storage engine implementation
Date Wed, 23 Jan 2013 03:28:12 GMT

    [ https://issues.apache.org/jira/browse/DRILL-16?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13560363#comment-13560363

Brian ONeill commented on DRILL-16:

I'm a big fan of GORA, but in my brief experience, the abstraction shields you from some of
the lower level operations that are available.  (and that is increasingly the case with things
like native support for collections: sets, maps, lists, etc.)   

As far as abstractions go, we could probably get a light-weight integration with Cassandra
going quickly using Kundera, which implements JPA/JPQL for Cassandra.  If the user didn't
need anything Cassandra specific (to optimize the Drill queries), then they could probably
just plug Kundera in and treat Cassandra like any other JDBC/relational store. (with CQL3,
the JPQL is almost equivalent in power to the direct NoSQL interactions)

If we want the data locality / topology information called out in DRILL-13, we'll probably
need to issue lower-level commands.  
Also, with Datastax advocating CQL and deprecating the old thrift interfaces, that might be
the better long term horse to hitch ourselves to.

(all IMHO)
> Cassandra storage engine implementation
> ---------------------------------------
>                 Key: DRILL-16
>                 URL: https://issues.apache.org/jira/browse/DRILL-16
>             Project: Apache Drill
>          Issue Type: Wish
>            Reporter: Gary Dusbabek
>            Priority: Minor
>              Labels: cassandra, drill, storage-engine

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

View raw message