A CPU and GPU Heterogeneous Processing of Multimedia Data by using OpenCL

1 "This paper is being submitted as a poster". Abstract In recent times, it has become possible to parallelize many multimedia applications using multicore platforms such as CPUs and GPUs. In this paper, we propose a parallel processing approach for a multimedia application by using both the CPU and GPU. Instead of distributing the parallelizable workload to either the CPU or GPU, we distribute the workload simultaneously to both by using OpenCL. Based on our experimental results, using both the CPU and GPU, we confirm that the proposed parallel processing approach provides better performance than typical parallel processing approaches on account of maximal utilization of the given resources.

[1]  Béatrice Pesquet-Popescu,et al.  OpenCL implementation of motion estimation for cloud video processing , 2011, 2011 IEEE 13th International Workshop on Multimedia Signal Processing.

[2]  Jian Cao,et al.  GPU Accelerated Target Tracking Method , 2011 .

[3]  Ivan Zelinka,et al.  GPU Based Enhanced Differential Evolution Algorithm: A Comparison between CUDA and OpenCL , 2013, Handbook of Optimization.

[4]  John E. Stone,et al.  OpenCL: A Parallel Programming Standard for Heterogeneous Computing Systems , 2010, Computing in Science & Engineering.