Parallelization clonal selection algorithm with MPI.NET for optimization problem

The idea of immune system as a computing inspiration has given rise to the Artificial Immune System (AIS). AIS has contributed in the field of optimization of complex issues, one of which is a clonal selection algorithm (CSA) for solving the Optimization Problem such as Traveling Salesperson Problems (TSP). Parallel characteristic inherently possessed by the immune system, provided an opportunity to give parallel computing to achieve better computational performance. The study resulted in two parallel models for the clonal selection algorithm. First model is a master-slave model, there is a master process that controls all communication. In the second model, all processes equivalent and communicate with each other. Both models are prepared to be built in system development with parallel environment using Visual C# language with MPI.NET framework. For both datasets, experiment gave consistent results. Model 1 is superior in getting the best-tour’s cost, although obtained with longer execution time, compared with Model 2. However, Model 2 is superior in less execution time needed.