SAMbA: Extending Apache Spark for Scientific Computational Experiments

This repository presents a quick start guide for SAMbA, an extension over Apache Spark, which aims at running black-box native programs that handle raw data files. Furthermore, this engine collects, stores, and query provenance and domain-specific data that were manipulated during the execution of scientific applications. With respect to the analytical capabilities, SAMbA provides runtime dataflow analysis based on the provenance traces.