Today, in the digitized satellite image domain, the needs for high dimension images increase considerably. To transmit or to stock such images (more than 6000 by 6000 pixels), we need to reduce their data volume and so we have to use image compression technics. I n most cases, these operations have to be processed in Real-Time. The large amount of computations required by classical image compression algorithms prohibits the use of common sequential processors.To solve this problem, CEA in collaboration with CNES has tried to define the best suited architecture for the image compression. In order to achieve this aim, we developed and evaluated a new parallel image compression algorithm for general purpose parallel computers using data-parallelism.The purpose of this paper is to present this new parallel image compression algorithm. W e present implementation results on several parallel computers. W e also examine load balancing and data mapping problems. W e end by defining optimal characteristics of the parallel machine for Real-Time image compression.

No abstract

The image processing applications require both computing and input/output power. The GFLOPS project's aim is to develop a parallel architecture as well as its software environment to implement those applications efficiently. This goal can be achieved only with a real collaboration among the architecture, the compiler and the programming language. This paper investigates the C/ / on global address space architectures. The main advantage of our paradigm is lhat it allows a unique framework to express both data and control parallelism. We will first present the structure of the GFLOPS machine used to implement this language. The C// parallel language will be presented in the next section, and finally we will evaluate the effectiveness of the mechanisms incorporated in the architecture to implement the high level C// structures.

No abstract

No abstract