Integrate Spark with MapR Streams

Integrate Spark with MapR Steams to enable Spark to query MapR Streams for new messages at a given interval, process any new messages that are available, and also publish messages into MapR Streams.

As of Spark 1.5.2-1602, you can use Spark to access MapR Streams through Spark’s receiver-less, direct approach. The integration between Spark 1.5.2 and MapR Streams is a beta feature.

For more information, see the Apache Spark documentation.

Note: Before you integrate Spark with MapR Streams, verify that the Streams Client is install on all Spark nodes. For more information, see Installing Spark on YARN or Installing Spark Standalone.