[
https://issues.apache.org/jira/browse/PHOENIX-971?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14238145#comment-14238145
]
Nick Dimiduk commented on PHOENIX-971:
--------------------------------------
[~jamestaylor]
bq. What kind of new dependencies would that require? How light/heavy would the
client-side be?
If we implement the thrift interface entirely on our own, we include whatever
dependencies we need to implement the service layer -- nothing HBase doesn't
already use, I imagine. Client would be a simple thrift client, nothing more.
bq. Another option would be to use the Apache Calcite Avatica RPC server
(http://s.apache.org/wwd).
I looked in Calcite for an RPC server but didn't find one. Thanks for the link.
I think pushing the implementation to Calcite makes a lot of sense. Hive has
invested a lot into HS2/beeline, I'm not excited to repeat that process. The
benefit of HS2 thrift over Avatica is maturity. There's already tooling for
non-Java languages to use HS2.
> Query server
> ------------
>
> Key: PHOENIX-971
> URL: https://issues.apache.org/jira/browse/PHOENIX-971
> Project: Phoenix
> Issue Type: New Feature
> Reporter: Andrew Purtell
>
> Host the JDBC driver in a query server process that can be deployed as a
> middle tier between lighter weight clients and Phoenix+HBase. This would
> serve a similar optional role in Phoenix deployments as the
> [HiveServer2|https://cwiki.apache.org/confluence/display/Hive/Setting+Up+HiveServer2]
> does in Hive deploys.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
