Generalization of orthogonal multiprocessor for massively parallel computation
- 6 January 2003
- proceedings article
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 391-398
- https://doi.org/10.1109/fmpc.1988.47459
Abstract
The authors present a multidimensional orthogonal multiprocessor (OMP) conceived for massively parallel computation. The OMP compares favorably with the generalized hypercube and spanning-bus hypercube. All three architectures extend the original concept of binary n-cube. Among the three, the OMP requires equal or less hardware and yet presents some interesting application potentials in massively parallel computation. The orthogonal architecture is a viable alternative to the conventional single-instruction multiple-data (SIMD) array computers which use distributed local memories. A detailed description of binary and k-ary OMP architecture is provided. Interprocessor communication strategy and network characteristics are presented. Potential applications and conclusions are includedKeywords
This publication has 5 references indexed in Scilit:
- Analysis and applications of the orthogonal access multiprocessorJournal of Parallel and Distributed Computing, 1989
- An orthogonal multiprocessor for parallel scientific computationsIEEE Transactions on Computers, 1989
- Generalized Hypercube and Hyperbus Structures for a Computer NetworkIEEE Transactions on Computers, 1984
- The ETH-Multiprocessor Empress: A Dynamically Configurable MIMD SystemIEEE Transactions on Computers, 1982
- Communication Structures for Large Networks of MicrocomputersIEEE Transactions on Computers, 1981