Proceedings of the 22nd International Conference on Parallel Architectures and Compilation Techniques (1996)
Oct. 20, 1996 to Oct. 23, 1996
F. Vivien , Lab. LIP, Ecole Normale Superieure de Lyon, France
A. Darte , Lab. LIP, Ecole Normale Superieure de Lyon, France
Abstract: This paper proposes an optimal algorithm for detecting fine or medium grain parallelism in nested loops whose dependences are described by an approximation of distance vectors by polyhedra. In particular it is optimal for direction vectors, which generalizes Wolf and Lam's algorithm (1991) to the case of several statements. It relies on a dependence uniformization process and an parallelization techniques related to system of uniform recurrence equations.
parallel algorithms; optimal fine grain parallelism detection; optimal medium grain parallelism detection; polyhedral reduced dependence graphs; optimal algorithm; nested loops; distance vectors; polyhedra; direction vectors; parallelization techniques; uniform recurrence equations
F. Vivien, A. Darte, "Optimal Fine and Medium Grain Parallelism Detection in Polyhedral Reduced Dependence Graphs", Proceedings of the 22nd International Conference on Parallel Architectures and Compilation Techniques, vol. 00, no. , pp. 0281, 1996, doi:10.1109/PACT.1996.552676