|Chapter title||A synthesis method of LSGP partitioning for given-shape regular arrays|
|Authors||Megson, G.M. and Chen, X.|
This paper presents a method to partition and map a computational polytope onto processor arrays. Based on the theoretical framework of an existing LSGP method, a systematic design procedure is proposed which constructs an activity matrix, proposed by Darte, according to the shapes of the computational polytope and the processor array and derives a valid timing vector. By this method the given-shape mapping can be achieved with neither difficulty nor exhausted computations by removing the need to compute HNFs.
|Keywords||Data structures, parallel algorithms, systolic arrays, LSGP partitioning, activity matrix, computational polytope given-shape regular arrays, locally sequential globally parallel, processor arrays, timing vector|
|Book title||Proceedings of the 9th International Parallel Processing Symposium, April 25-28, 1995, Santa Barbara, California|
|Place of publication||Los Alamitos, USA|
|Digital Object Identifier (DOI)||https://doi.org/10.1109/IPPS.1995.395938|