Distributed genetic algorithm on GraphX

In this paper we explore the application of a recent breed of distributed systems, graph processing frameworks in particular, to solving complex research problems. These frameworks are designed to take full advantage of today’s abundant resources with their inherent distributed computing functionalities. Abstraction of many technical details, such as networking and coordination of multiple compute nodes is a desirable feature provided by these graph-processing frameworks. While these frameworks are largely used to process and analyse the web graphs and social networks, their capacity is not limited to this direct application. This paper is based on design and implementation of a genetic algorithm (GA) using a graph processing tool, GraphX for the task scheduling problem as a case study. Our experimental results show that GraphX can significantly aid in devising distributed solutions for complex problems.7 page(s