high performance computing on graphics processing units: hgpu.org

Data compression is an important area of information and communication technologies it seeks to reduce the number of bits used to store or transmit information. It will efficiently utilizes the memory spaces and allows to transmit data within a limited bandwidth. Most compression process is achieved by removing data redundancy while preserving information content. Data compression algorithms exploit some characteristics to make the compressed data smaller than the original data. Every data compression process is working with well defined algorithm. Data compression on graphics processors (GPUs) has become an effective approach to improve the performance of main memory. CUDA is a parallel computing platform and programming model invented by NVIDIA. It enables dramatic increases in computing performance with graphics processing unit (GPU).Data compression algorithms on CUDA provides better compression process. In this paper, we implement the most power full algorithm LZW on CUDA architecture. Due to the parallel characteristics of GPU, compression process time is very less than the CPU environment.