Modelling the 2-D FCT on a Multiprocessor System (original) (raw)
Algorithms and Parallel VLSI Architectures III, 1995
Abstract
ABSTRACT. This paper describes the parallel implementation of the 2-D Fast Cosine Transform (FCT). The examined algorithms are: Row-Column (RC) FCT and Vector-Radix (VR) FCT. A SUN SPARC 10 with 4 processors and shared-memory is considered. It is observed that the parallel approach provides considerable gain for block sizes of more than 128x128 pixels. For smaller sizes there is no speed-up: by using more than one processor, the execution times are larger than on one processor due to the time required for the ...
Jan Cornelis hasn't uploaded this paper.
Create a free Academia account to let Jan Cornelis know you want this paper to be uploaded.
Ask for this paper to be uploaded.