Knime BigData Local Environment & findspark

Question

I'm just wondering: is possible to access the BigData Environment Node launched in Knime?

I have a node in a Knime workflow. From outside Knime, I wrote a Python script using findspark but ... it doesn't seem to detect it.

Any example or proposal to do that?

Thank you in advance

You mean you would like to access big data environment created by this node or? — ipazin, Dec 18 '19 at 17:50
Yes @ipazin, this is the case. I would like to access this hadoop node (provided by Knime) from external applications (python script for instance). Do you know url, ports in order to be able to connect (HDFS, Hive etc.) Thank you in advance? — Julián Gómez Cuadrado, Feb 21 '20 at 07:49

score 0 · Accepted Answer · answered Mar 06 '20 at 12:19

0

Hive is accessible while Spark and HDFS are not. You can connect to Hive via JDBC with port which is randomized upon node execution. Information about it should be in the first output port - DB Session.

answered Mar 06 '20 at 12:19

ipazin

102
7

Knime BigData Local Environment & findspark

1 Answers1