Self-organizing migration algorithm on GPU with CUDA

Abstract

A modification of Self-organizing migration algorithm for general-purpose computing on graphics processing units is proposed in this paper. The algorithm is implemented in C++ with its core parts in c-CUDA. Its implementation details and performance are evaluated and compared to previous, pure C++ version of algorithm. 6 commonly used artificial test functions are used to test the performance. The test results clearly show significant speed gains without a compromise in convergence quality.