RT Conference Proceedings
T1 CUVLE: Variable-Length Encoding on CUDA
A1 Fuentes-Alventosa, Antonio
A1 Gómez-Luna, Juan
A1 González-Linares, José María
A1 Guil-Mata, Nicolás
K1 Arquitectura de ordenadores
AB Data compression is the process of representinginformation in a compact form, in order to reduce the storagerequirements and, hence, communication bandwidth. It has beenone of the critical enabling technologies for the ongoing digitalmultimedia revolution for decades. In the variable-lengthencoding (VLE) compression method, most frequently occurringsymbols are replaced by codes with shorter lengths. As it is acommon strategy in many compression applications, efficientparallel implementations of VLE are very desirable. In this paperwe present CUVLE, a GPU implementation of VLE on CUDA.Our approach is on average more than 20 and 2 times faster thanthe corresponding CPU serial implementation and the onlyknown state-of-the-art GPU implementation, respectively.
YR 2014
FD 2014-10-14
LK http://hdl.handle.net/10630/8224
UL http://hdl.handle.net/10630/8224
LA eng
NO Junta de Andalucía, TIC-1692. Universidad de Málaga. Campus de Excelencia Internacional Andalucía Tech
DS RIUMA. Repositorio Institucional de la Universidad de Málaga
RD 10 mar 2026