High Performance Analytics Toolkit (HPAT) is a compiler-based framework for big data analytics and machine learning on cluster/cloud environments that is both easy to use and extremely fast; it is orders of magnitude faster than alternatives like Apache Spark.
HPAT automatically or semi-automatically parallelizes analytics tasks written in Numpy and pandas and generates efficient MPI code using Numba and LLVM. These academic papers describe the underlying methods in HPAT:
HPAT's documentation can be found here.