Forum

[tspann]

I am going to try to install on my OpenStack Centos 7.2 HDP 2.5 cluster to try it out.

Also to use for my meetup
http://meetup.com/futureofdata-princeton

Any tips or suggestions?

Can I run queries in Hive? Spark SQL? Zeppelin? hive CLI? beeline?

[markc]

Reading http://www.kognitio.com/forums/Getting% ... Hadoop.pdf is a good starting point.

Then you can download the Kognitio client tools and run e.g. Kognitio Console against the system to submit SQL. The default is that the edge node you initiate the Kognitio YARN application from will listen on port 6550 for ODBC connections. So if you install the Kognitio clients and create a DSN with the edge node's IP address specified for connections, you can then run SQL. You could also use other third party tools like Tableau, Microstrategy, etc. to connect to your Kognitio application and query data that is stored in the Hadoop cluster.

The general Kognitio documentation explains how to e.g. create external tables which allow you to access data stored in HDFS. Ideally you'd create in-memory images of these to allow you to have best performance - see the general documentation on create view images here, for example.

Mark.