Abstract
A new class of interconnection networks is proposed for interconnecting the processors of a general purpose parallel computer which is based on the hierarchical application of a complete graph compound. The systematic construction of this new class of interconnection networks, RCC, is shown and its properties are derived and are compared favorably to other interconnection networks. A specific instance of this class, RCC-CUBE, is shown to have desirable network properties such as small diameter, small degree, high density, and high bandwidth. The hardware cost and physical time performance are estimated for RCC-CUBE and compared to those of the hypercube and the 2-D mesh demonstrating an overall cost effectiveness for RCC-CUBE. Thus, the RCC-CUBE appears to be a good candidate for next generation massively parallel computer systems.
| Original language | English |
|---|---|
| Pages | 121-130 |
| Number of pages | 10 |
| Publication status | Published - 1995 |
| Externally published | Yes |
| Event | Proceedings of the IEEE 1st International Conference on Algorithms and Architectures for Parallel Processing. Part 1 (of 2) - Brisbane, Aust Duration: 19 Apr 1995 → 21 Apr 1995 |
Conference
| Conference | Proceedings of the IEEE 1st International Conference on Algorithms and Architectures for Parallel Processing. Part 1 (of 2) |
|---|---|
| City | Brisbane, Aust |
| Period | 19/04/95 → 21/04/95 |