Title :
Hybrid technology multithreaded architecture
Author :
Gao, Guang ; Likharev, Konstantin K. ; Messina, Paul C. ; Sterling, Thomas L.
Author_Institution :
Delaware Univ., Newark, DE, USA
Abstract :
The objective of the hybrid technology multithreaded architecture (HTMT) research is to determine the feasibility and structure of a parallel architecture integrating the combined capabilities of semiconductor, superconductor, and optical technologies. This mix of technologies may yield operational attributes superior to those based solely on semiconductor technology. The authors present a brief overview of the crucial technology-the superconducting rapid single-flux quantum logic (RSFQ) technology, as well as the HTMT multithreaded program execution models and architecture. The HTMT approach exploits key emerging devices. Specifically, computational performance can be dramatically improved through recent advances in RSFQ technology, making 100 GHz clock rates feasible in the near future. Memory capacity may be drastically increased through a new memory hierarchy merging superconductor memory, advanced semiconductor high-density memory and future optical 3D holographic storage. Interconnection bandwidth will be greatly enhanced by means of optical networks with very high bandwidth
Keywords :
holographic storage; parallel architectures; semiconductor storage; superconducting logic circuits; 100 GHz; advanced semiconductor high-density memory; clock rates; computational performance; hybrid technology multithreaded architecture; interconnection bandwidth; memory capacity; memory hierarchy; multithreaded program execution models; optical 3D holographic storage; optical networks; optical technology; parallel architecture; semiconductor technology; superconducting rapid single-flux quantum logic technology; superconductor memory; superconductor technology; Bandwidth; Clocks; Computer architecture; Holographic optical components; Holography; Integrated optics; Logic devices; Merging; Parallel architectures; Superconducting logic circuits;
Conference_Titel :
Frontiers of Massively Parallel Computing, 1996. Proceedings Frontiers '96., Sixth Symposium on the
Conference_Location :
Annapolis, MD
Print_ISBN :
0-8186-7551-9
DOI :
10.1109/FMPC.1996.558066