Exploiting multilevel parallelism within modern microprocessors: DWT as a case study

Authors:
C. Tenllado;C. Garcia;M. Prieto;L. Piñuel;F. Tirado
Affiliations:
Departamento de Arquitectura de Computadores y Automática, Universidad Complutense, Madrid, Spain;Departamento de Arquitectura de Computadores y Automática, Universidad Complutense, Madrid, Spain;Departamento de Arquitectura de Computadores y Automática, Universidad Complutense, Madrid, Spain;Departamento de Arquitectura de Computadores y Automática, Universidad Complutense, Madrid, Spain;Departamento de Arquitectura de Computadores y Automática, Universidad Complutense, Madrid, Spain
Venue:
VECPAR'04 Proceedings of the 6th international conference on High Performance Computing for Computational Science
Year:
2004

Citing 10
Cited 0

Simultaneous multithreading: maximizing on-chip parallelism

ISCA '95 Proceedings of the 22nd annual international symposium on Computer architecture
JPEG 2000: Image Compression Fundamentals, Standards and Practice

JPEG 2000: Image Compression Fundamentals, Standards and Practice
AltiVec Extension to PowerPC Accelerates Media Processing

IEEE Micro
A survey of processors with explicit multithreading

ACM Computing Surveys (CSUR)
-D Wavelet Transform Enhancement on General-Purpose Microprocessors: Memory Hierarchy and SIMD Parallelism Exploitation

HiPC '02 Proceedings of the 9th International Conference on High Performance Computing
Vectorization of the 2D Wavelet Lifting Transform Using SIMD Extensions

IPDPS '03 Proceedings of the 17th International Symposium on Parallel and Distributed Processing
Kernel korner: CPU affinity

Linux Journal
Billion-Transistor Architectures: There and Back Again

Computer
IBM Power5 Chip: A Dual-Core Multithreaded Processor

IEEE Micro
Optimal memory organization for scalable texture codecs in MPEG-4

IEEE Transactions on Circuits and Systems for Video Technology

Quantified Score

Hi-index	0.00

Visualization

Abstract

Simultaneous multithreading (SMT) is being incorporated into modern superscalar microprocessors, allowing several independent threads to issue instructions to the functional units in a single cycle. Effective use of the SMT can hide the inefficiencies caused by long operation latencies, thereby yielding a better utilization of the processor's resources. In this paper we explore techniques to efficiently exploit this capability and its interaction with short-vector processing. We put special emphasis on the differences in algorithm tuning between SMT architectures and shared memory symmetric multiprocessors. As a case study we have chosen the well known Discrete Wavelet Transform (DWT), a central-piece in some image and video coding standards such as MPEG-4 or JPEG-2000.