Processes can now be executed in the background of RapidMiner Studio while you continue working on the design of your process. Note this only applies to users of RapidMiner Studio Large.
Extension developers will now have the option to use our concurrency API to develop operators with parallel execution and an improved performance.
The new SparkRM feature opens new use cases including text processing, series forecasting, process mining, and more. With SparkRM, any operator or process in RapidMiner Studio can be run in parallel on a Hadoop cluster.
The number of operators that you can use in Hadoop increases from 70 to over 1500.
Hadoop’s impersonation functionality is now supported, allowing simpler and more secure connections in multi-user Hadoop clusters.
Two things that are getting me excited. One is the new paralellization framework. More speed = more bettah and SparkRM. It took a lot of fortitude on my part not to say anything about this before the release but SparkRM opens up all extensions, operators, and RapidMiner goodness from your desktop to the Hadoop Cluster. Everything runs on HADOOP!
There's going to be a webinar on Feb 28th to showcase all the new changes. Signup for it here.