Optimizing LZSS compression on GPGPUs. Ozsoy, A., Swany, M., & Chauhan, A. Future Generation Computer Systems, 2014.
abstract   bibtex   
In this paper, we present an algorithm and provide design improvements needed to port the serial Lempel-Ziv-Storer-Szymanski (LZSS), lossless data compression algorithm, to a parallelized version suitable for general purpose graphic processor units (GPGPU), specifically for NVIDIA's CUDA Framework. The two main stages of the algorithm, substring matching and encoding, are studied in detail to fit into the GPU architecture. We conducted detailed analysis of our performance results and compared them to serial and parallel CPU implementations of LZSS algorithm. We also benchmarked our algorithm in comparison with well known, widely used programs: GZIP and ZLIB. We achieved up to 34× better throughput than the serial CPU implementation of LZSS algorithm and up to 2.21× better than the parallelized version. © 2013 Elsevier B.V. All rights reserved.
@article{
 title = {Optimizing LZSS compression on GPGPUs},
 type = {article},
 year = {2014},
 identifiers = {[object Object]},
 volume = {30},
 id = {65f19ac8-0b67-348c-af6e-e9efcf6d2709},
 created = {2019-10-01T17:20:58.947Z},
 file_attached = {false},
 profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
 last_modified = {2019-10-01T17:23:58.379Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Ozsoy2014},
 folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
 private_publication = {false},
 abstract = {In this paper, we present an algorithm and provide design improvements needed to port the serial Lempel-Ziv-Storer-Szymanski (LZSS), lossless data compression algorithm, to a parallelized version suitable for general purpose graphic processor units (GPGPU), specifically for NVIDIA's CUDA Framework. The two main stages of the algorithm, substring matching and encoding, are studied in detail to fit into the GPU architecture. We conducted detailed analysis of our performance results and compared them to serial and parallel CPU implementations of LZSS algorithm. We also benchmarked our algorithm in comparison with well known, widely used programs: GZIP and ZLIB. We achieved up to 34× better throughput than the serial CPU implementation of LZSS algorithm and up to 2.21× better than the parallelized version. © 2013 Elsevier B.V. All rights reserved.},
 bibtype = {article},
 author = {Ozsoy, A. and Swany, M. and Chauhan, A.},
 journal = {Future Generation Computer Systems},
 number = {1}
}

Downloads: 0