Title :
Mapping nested loop algorithms into multidimensional systolic arrays
Author :
Lee, Pei-Zong ; Kedem, Zvi Meir
Author_Institution :
Courant Inst. of Math. Sci., New York Univ., NY, USA
fDate :
1/1/1990 12:00:00 AM
Abstract :
Consideration is given to transforming depth p-nested for loop algorithms into q-dimensional systolic VLSI arrays where 1⩽q⩽p-1. Previously, there existed complete characterizations of correct transformation only for the cases where q=p-1 or q=1. This gap is filled by giving formal necessary and sufficient conditions for correct transformation of a p-nested loop algorithm into a q-dimensional systolic array for any q, 1⩽q⩽p-1. Practical methods are presented. The techniques developed are applied to the automatic design of special purpose and programmable systolic arrays. The results also contribute toward automatic compilation onto more general purpose programmable arrays. Synthesis of linear and planar systolic array implementations for a three-dimensional cube-graph algorithm and a reindexed Warshall-Floyd path-finding algorithm are used to illustrate the method
Keywords :
cellular arrays; graph theory; matrix algebra; parallel algorithms; algorithm transformations; automatic compilation; correct transformation; data dependence; general purpose programmable arrays; matrix multiplication; multidimensional systolic arrays; necessary conditions; nested loop algorithms; parallel processing; planar systolic array implementations; programmable systolic arrays; reindexed Warshall-Floyd path-finding algorithm; sufficient conditions; three-dimensional cube-graph algorithm; Computer science; History; Information science; Multidimensional systems; Parallel processing; Sufficient conditions; Systolic arrays; Transmission line matrix methods; Vectors; Very large scale integration;
Journal_Title :
Parallel and Distributed Systems, IEEE Transactions on