Parallel coral reef algorithm for solving JSP on Spark

A high-performance scheduling algorithm, called MapReduce coral reef (MRCR), is presented in this paper. The basic idea of the proposed algorithm is to apply the MapReduce programming model of Spark to the coral reef optimization algorithm to speed up its response time. This means that MRCR will distribute its solutions to as many cluster nodes as it needs for calculating the objective value of solutions using the MapReduce technology. To evaluate the performance of the proposed algorithm, several benchmarks are used with different algorithms. Our experimental results show that the proposed algorithm is able to speed up the response time of the coral reef optimization algorithm in solving complex optimization problems, especially when the size of benchmarks is large.