TY - JOUR ID - 2020 TI - A Distributed Sailfish Optimizer Based on Multi-Agent Systems for Solving Non-Convex and Scalable Optimization Problems Implemented on GPU JO - Journal of AI and Data Mining JA - JADM LA - en SN - 2322-5211 AU - Shadravan, S. AU - Naji, H. AU - Khatibi, V. AD - Department of Computer Engineering, Kerman Branch, Islamic Azad University, Kerman, Iran AD - Department of Computer Engineering, Graduate University of Advanced Technology, Kerman, Iran. AD - Department of Computer Engineering, Kerman Branch, Islamic Azad University, Bardsir, Iran. Y1 - 2021 PY - 2021 VL - 9 IS - 1 SP - 59 EP - 71 KW - SailFish Optimizer (SFO) KW - Multi agent system KW - parallel processing KW - shared memory KW - Graphic processing units DO - 10.22044/jadm.2020.9389.2075 N2 - The SailFish Optimizer (SFO) is a metaheuristic algorithm inspired by a group of hunting sailfish that alternates their attacks on group of prey. The SFO algorithm takes advantage of using a simple method for providing the dynamic balance between exploration and exploitation phases, creating the swarm diversity, avoiding local optima, and guaranteeing high convergence speed. Nowadays, multi agent systems and metaheuristic algorithms can provide high performance solutions for solving combinatorial optimization problems. These methods provide a prominent approach to reduce the execution time and improve of the solution quality. In this paper, we elaborate a multi agent based and distributed method for sailfish optimizer (DSFO), which improves the execution time and speedup of the algorithm while maintaining the results of optimization in high quality. The Graphics Processing Units (GPUs) using Compute Uniļ¬ed Device Architecture (CUDA) are used for the massive computation requirements in this approach. In depth of the study, we present the implementation details and performance observations of DSFO algorithm. Also, a comparative study of distributed and sequential SFO is performed on a set of standard benchmark optimization functions. Moreover, the execution time of distributed SFO is compared with other parallel algorithms to show the speed of the proposed algorithm for solving unconstrained optimization problems. The final results indicate that the proposed method is executed about maximum 14 times faster than other parallel algorithms and shows the ability of DSFO for solving non-separable, non-convex and scalable optimization problems. UR - https://jad.shahroodut.ac.ir/article_2020.html L1 - https://jad.shahroodut.ac.ir/article_2020_e2dc619f13c08e5a7f8951507c7ade2f.pdf ER -