Hadoop Mapreduce to Parallelize Social Spider Optimization

Social Spider Optimization (SSO) has received attention in many research fields and real-world applications for solving optimization problems. Factor that affects the performance of SSO is its imbalance of exploration and exploitation. Its ability of the exploration in a multi-dimensional solution pace increases the execution time quite significantly. To reduce the execution time, parallel implementation of SSO should be implemented. In this paper, we implement and compare the parallel implementation of SSO using two different parallelization techniques using MapReduce programming, 1) all nodes in the cluster work on the same population, and 2) each node in cluster has its own population. Both parallel implementations are compared based on performance and speedup. Parallel implementation of the SSO algorithm makes the algorithm faster in case of both low and high dimensional datasets.