high performance computing on graphics processing units: hgpu.org

The main contribution of this paper is to present a parallel algorithm for LZW decompression and to implement it in a CUDA-enabled GPU. Since sequential LZW decompression creates a dictionary table by reading codes in a compressed file one by one, its parallelization is not an easy task. We first present a parallel LZW decompression algorithm on the CREW-PRAM. We then go on to present an efficient implementation of this parallel algorithm on a GPU. The experimental results show that our parallel LZW decompression on GeForce GTX 980 runs up to 69.4 times faster than sequential LZW decompression on a single CPU. We also show a scenario that parallel LZW decompression on a GPU can be used for accelerating big data applications.