Evaluating new approaches of big data analytics frameworks

Abstract

The big data topic will be one of the leading growth markets in information technology in the next years. One problem in this area is the efficient computation of huge data volumes, especially for complex algorithms in data mining and machine learning tasks. This paper discuss new processing frameworks for big and smart data in distributed environments and presents a benchmark between two frameworks - Apache Flink and Apache Spark - based on a mixed workload with algorithms from different analytic areas with different real-world datasets.