An efficient associative processor for two primitive numerical operations employed in computer simulation applications

Fast and efficient computers are in high demand in many application areas of computer simulation, due to the large-scale computations that are often involved. Parallel processing computers are needed to meet these computational demands. Therefore, efficient high performance computing architectures and parallel algorithms that fully exploit their capabilities should be an aim of continuous research. Recent innovations in VLSI technology have enabled the design of powerful massively parallel computers, such as the associative processor. In this paper, we introduce an associative processor with a single-stage hypercube interconnection among memory words, and with the word-network interface providing routing function and small buffering (two field-size memory buffers, one for transmission and one for reception). We also provide additional interword connectivity via bidirectional shifting between adjacent words. An efficient low-cost system results that allows parallel data transfers and achieves dramatic performance improvements for key numerical applications in computer simulation, as we demonstrate introducing efficient algorithms for matrix multiplication and for one dimensional Fast Fourier Transform (FFT), achieving linear and logarithmic complexity, respectively.