Automatic Data Mapping of Signal Processing Applications

Corinne Ancourt, Denis Barthou, Christophe GUETTIER, Francois Irigoin, Bertrand Jeannet, Jean Jourdan, Juliette Mattioli

Abstract: This paper presents a technique to map automatically a complete DSP application onto a parallel machine with distributed memory. Unlike other applications where coarse or medium grain scheduling techniques can be used, DSP applications integrate several thousand of tasks and hence necessitate fine grain considerations. Moreover, finding an effective mapping imperatively requires to take into account both architectural resources and real time constraints. The main contribution of this paper is to show it is possible to handle and to solve data partitioning, and fine-grain scheduling under the above operatinal constraints using Concurrent Constraints Logic Programming (CCLP). Our concurrent resolution technique undertaking linear and non linear constraints takes advantage of the special features of signal processing applications and provides a solution equivalent to a manual solution for the representative Panoramic Analysis (PA) application.