Parallel Symbiotic Organisms Search Algorithm

Symbiotic organisms search algorithm is a population-based evolutionary optimization technique that is motivated by the simulation of social behaviour that emanates from the symbiosis relationship amongst organisms in an ecosystem. It is a popular global search swarm intelligence metaheuristic that is widely being used in conjunction with several other algorithms in different fields of study. Fascinatingly, the algorithm has also been shown to have the capability of optimizing several NP-hard problems in both continuous and binary search spaces. More so, because most of the modern day real-world computational problems requires machines with high processing power and improved optimization techniques, it is important to find ways to improve the speedup of the optimization process of this algorithm, as the complexity of the problems increase. Therefore, this paper explores the possibility of improving the optimization speedup and performance of the symbiotic organisms search algorithm through parallelization methods. The proposed parallelization procedure is implemented using OpenMP on a shared memory architecture and evaluated on a set of twenty mathematical test problems. The computational results of the parallel symbiotic organisms search algorithm was compared to its serial counterpart using a measure of run-time complexity.