The rapidly changing capabilities of modern graphics processing units (GPUs) give developers the opportunity to implement parallel programming techniques, or even combine the traditional (single process) programming with parallel programming. As the computing is shifting from central processing (on the CPU) to co-processing (on the CPU and GPU), platforms have emerged and gave the developers opportunity to experiment. In this work, a parallel solution of the modified subset sum algorithm is implemented using OpenCL, and the obtained speedup is compared to the CPU version and other parallel implementations of the problem.
Parallel computing, Subset Sum, OpenCL, GPU
© Association for Information and Communication Technologies www.ict-act.org Developed by tdelev