Overview of Parallel Platforms for Common High Performance Computing
Tomáš Frýza, Jitka Svobodová, Filip Adamec, Roman Maršálek, Jan Prokopec · Brno University of Technology Digital Library (Brno University of Technology) · 2012
Abstract. The paper deals with various parallel platforms used for high performance computing in the signal process-ing domain. More precisely, the methods exploiting the mul-ticore central processing units such as message passing in-terface and OpenMP are taken into account. The properties of the programming methods are experimentally proved in the application of a fast Fourier transform and a discrete cosine transform and they are compared with the possibili-ties of MATLAB’s built-in functions and Texas Instruments digital signal processors with very long instruction word ar-chitectures. New FFT and DCT implementations were pro-posed and tested. The implementation phase was compared with CPU based computing methods and with possibilities of the Texas Instruments digital signal processing library on C6747 floating-point DSPs. The optimal combination of computing methods in the signal processing domain and new, fast routines ’ implementation is proposed as well. Keywords Digital signal processing, fast Fourier transforms, dis-crete cosine transforms, parallel programming, high performance computing, message passing interface, OpenMP, MATLAB, digital signal processors, opti-mization.