Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

On partitioning and mapping for hypercube computing

Journal Article · · International Journal of Parallel Programming; (USA)
DOI:https://doi.org/10.1007/BF01407815· OSTI ID:6389121
 [1];  [2]
  1. Michigan State Univ., East Lansing (USA)
  2. New Jersey Institute of Technology, Newark (USA)
Designing efficient parallel algorithms in a message-based parallel computer should consider both time-space tradeoffs and computation-communication tradeoffs. In order to balance these tradeoffs and achieve the optimal performance of an algorithm, one has to consider various design parameters such as the number of processors required and the size of partitions. In this paper, the authors demonstrate that, for certain data parallel algorithms, it is possible to determine these design parameters analytically. To serve as a basis for the discussions that follow, a simple model for the NCUBE hypercube computer is introduced. Using this model, they use two examples, array summation and matrix multiplication, to illustrate how their performance can be modeled. By optimizing these expressions, one is able to determine optimal design parameters which arrive at efficient execution. Experiments on a 64-node NCUBE verified the accuracy of the analytic results and are used to further support the discussions.
OSTI ID:
6389121
Journal Information:
International Journal of Parallel Programming; (USA), Journal Name: International Journal of Parallel Programming; (USA) Vol. 17:6; ISSN IJPPE; ISSN 0885-7458
Country of Publication:
United States
Language:
English