Rheem goes open source
The Rheem project has now published its source code under the Apache License on GitHub.
Rheem is a cross-platform data analytics tool that allows users to formulate data analytics applications independent of any underlying platform. Given such an application, Rheem's optimizer automatically chooses (a combination of) appropriate platforms to run the application on and its execution component takes care of orchestrating the chosen platforms to complete the application's tasks. This approach frees users from tying their applications to concrete platforms and also permits for dynamic execution strategies based on variant circumstances, such as the size of datasets being analyzed.
The Rheem project originates from and is led by the QCRI data analytics group. The research and development for the open-sourced prototype has also been conducted to a great extent by Sebastian Kruse, while and after interning at QCRI. For more information, visit the Rheem webpage.