2026-06-01T01:54:41Zhttps://docta.ucm.es/rest/oai/request

oai:docta.ucm.es:20.500.14352/609002024-08-30T17:45:05Zcom_20.500.14352_14col_20.500.14352_21

Wavelet transform for large scale image processing on modern microprocessors Chaver Martínez, Daniel Ángel Tenllado Van Der Reijden, Christian Tomás Piñuel Moreno, Luis Tirado Fernández, José Francisco 004 Performance Informática (Informática) Programación de ordenadores (Informática) 1203.17 Informática 1203.23 Lenguajes de Programación © 2003 Springer Verlag. International Conference on High Performance Computing for Computational Science (VECPAR 2002) (5th. jun 26-28, 2002. Oporto, Portugal). In this paper we discuss several issues relevant to the vectorization of a 2-D Discrete Wavelet Transform on current microprocessors. Our research is based on previous studies about the efficient exploitation of the memory hierarchy, due to its tremendous impact on performance. We have extended this work with a more detailed analysis based on hardware performance counters and a study of vectorization, in particular, we have used the Intel Pentium SSE instruction set. Most of our optimizations are performed at source code level to allow automatic vectorization, though some compiler intrinsic functions have been introduced to enhance performance. Taking into account the abstraction at which the optimizations are performed, the results obtained on an Intel Pentium III microprocessor are quite satisfactory, even though further improvement can be obtained by a more extensive use of compiler intrinsics. Sección Deptal. de Arquitectura de Computadores y Automática (Físicas) Fac. de Ciencias Físicas TRUE pub 2023-06-20T21:10:49Z 2023-06-20T21:10:49Z 2003 book part https://hdl.handle.net/20.500.14352/60900 XXXX-XXXX eng open access application/pdf Springer-Verlag Berlin