Parallel JPEG2000 Image Coding on Multiprocessors
Peter Meerwald, Roland Norcen und Andreas Uhl
Proceedings of the International Parallel & Distributed Processing
Symposium, IPDPS '02
Fort Lauderdale, FL, USA
April 15-19, 2002
as BibTeX-file
conference program
Abstract
In this paper, we discuss how the two reference implementations of
the upcoming JPEG2000 image coding standard can be parallelized for the
execution on shared-memory multiprocessors.
By runtime analysis, we identify two major stages in
the coding process of JPEG2000 where parallelism can be exploited. We
present
techniques to exploit the parallelism within these two stages, and
speedup results obtained
on several hardware platforms. We focus on OpenMP as well as JAVA threads
for programming within
shared-memory environments.
Keywords: JPEG2000, wavelets, parallel, compression,
multi-processors, SMP, OpenMP
Links
Available in Acrobat (.pdf, zipped) und Postscript (.ps.gz, gzipped)
format.
Peter Meerwald, pmeerw@cosy.sbg.ac.at