6 billions, mono-bit operations per second : the cellular automata approach to the simulation of natural phenomenon, using a systolic GAPP array. Application to the diffusion

摘要

The « cellular automata » approch (see the references [8] to [17]) to theudsimulation of natural phenomenon leads up to computation problemsudessentially due ta the required computation power and to the data transferudbottleneck between processing elements and memories .udA real-time image processing system for low and medium level processingudthat we developped at LSIT, Strasbourg, enabled us to solve theseudproblems. A systolic SIMD array (GAPP of NCR) gaved us enoughudprocessing power to handle the iterative cellular automata method . In this paper, we will first describe our image processing systemud(MASYVE) and give the simulated diffusion model . We will then describeudthe management of data transfers . At last we will present the achievedudperformances, the limitations of the system and the perspectives .