TY - JOUR
T1 - A spanning multichannel linked hypercube
T2 - A gradually scalable optical interconnection network for massively parallel computing
AU - Louri, Ahmed
AU - Weech, Brent
AU - Neocleous, Costas
N1 - Funding Information:
puter engineering in 1988, the MS degree in computer engineering in 1984, both from the University of Southern California, and the Dip-lome D’Engenieur (Engineer Degree) in electrical engineering from the University of Science and Technology, Oran, Algeria, in 1982. He is cur-rently an associate professor of electrical and computer engineering at the University of Ari-zona and director of the Optical Networking and Parallel Processing Laboratory. His research interests include computer architecture, parallel processing, optical computing, and optical interconnects. Dr. Louri has published numerous journal and conference articles on the above topics. In 1991, he received the “Best Article of 1991 Award” from IEEE Micro. In 1988, he was the recipient of the U.S. National Science Foundation Research Initial Award. In 1994, he was the recipient of the Advanced Telecommunications Organization of Japan Fellowship, Ministry of Post and Telecommunications, Japan. In 1995, he was the recipient of the Centre Nationale de Recherche Scientifique (CNRS), France fellowship. In 1996, he was the recipient of the Japanese Society for the Promotion of Science fellowship. Prior to joining the University of Arizona, he worked as a researcher with the Computer Research Institute at the University of Southern California, where he conducted extensive research in parallel processing, multiprocessor system design, and optical computing. He has served as a member of the Technical Program Committee of several conferences, including OSA Topical Meeting on Optics in Computing, OSA/IEEE Conference on Massively Parallel Processors Using Optical Interconnects, among others. In 1996, he was the general chair of the Workshop on Optics in High-Performance Computing at Euro-Par ’96, Lyon, France.
Funding Information:
The authors would like to thank the anonymous referees for their constructive remarks. This work was supported by the U.S. National Science Foundation under Grant MIP 93-10082.
PY - 1998
Y1 - 1998
N2 - A new, scalable interconnection topology called the Spanning Multichannel Linked Hypercube (SMLH) is proposed. This proposed network is very suitable to massively parallel systems and is highly amenable to optical implementation. The SMLH uses the hypercube topology as a basic building block and connects such building blocks using two-dimensional multichannel links (similar to spanning buses). In doing so, the SMLH combines positive features of both the hypercube (small diameter, high connectivity, symmetry, simple routing, and fault tolerance) and the spanning bus hypercube (SBH) (constant node degree, scalability, and ease of physical implementation), while at the same time circumventing their disadvantages. The SMLH topology supports many communication patterns found in different classes of computation, such as bus-based, mesh-based, and tree-based problems, as well as hypercube-based problems. A very attractive feature of the SMLH network is its ability to support a large number of processors with the possibility of maintaining a constant degree and a constant diameter. Other positive features include symmetry, incremental scalability, and fault tolerance. It is shown that the SMLH network provides better average message distance, average traffic density, and queuing delay than many similar networks, including the binary hypercube, the SBH, etc. Additionally, the SMLH has comparable performance to other high-performance hypercubic networks, including the Generalized Hypercube and the Hypermesh. An optical implementation methodology is proposed for SMLH. The implementation methodology combines both the advantages of free space optics with those of wavelength division multiplexing techniques. A detailed analysis of the feasibility of the proposed network is also presented.
AB - A new, scalable interconnection topology called the Spanning Multichannel Linked Hypercube (SMLH) is proposed. This proposed network is very suitable to massively parallel systems and is highly amenable to optical implementation. The SMLH uses the hypercube topology as a basic building block and connects such building blocks using two-dimensional multichannel links (similar to spanning buses). In doing so, the SMLH combines positive features of both the hypercube (small diameter, high connectivity, symmetry, simple routing, and fault tolerance) and the spanning bus hypercube (SBH) (constant node degree, scalability, and ease of physical implementation), while at the same time circumventing their disadvantages. The SMLH topology supports many communication patterns found in different classes of computation, such as bus-based, mesh-based, and tree-based problems, as well as hypercube-based problems. A very attractive feature of the SMLH network is its ability to support a large number of processors with the possibility of maintaining a constant degree and a constant diameter. Other positive features include symmetry, incremental scalability, and fault tolerance. It is shown that the SMLH network provides better average message distance, average traffic density, and queuing delay than many similar networks, including the binary hypercube, the SBH, etc. Additionally, the SMLH has comparable performance to other high-performance hypercubic networks, including the Generalized Hypercube and the Hypermesh. An optical implementation methodology is proposed for SMLH. The implementation methodology combines both the advantages of free space optics with those of wavelength division multiplexing techniques. A detailed analysis of the feasibility of the proposed network is also presented.
KW - Interconnection networks
KW - Massively parallel processing
KW - Optical interconnects
KW - Product networks
KW - Scalability
KW - Wavelength division multiplexing
UR - http://www.scopus.com/inward/record.url?scp=0032074166&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0032074166&partnerID=8YFLogxK
U2 - 10.1109/71.679219
DO - 10.1109/71.679219
M3 - Article
AN - SCOPUS:0032074166
SN - 1045-9219
VL - 9
SP - 497
EP - 512
JO - IEEE Transactions on Parallel and Distributed Systems
JF - IEEE Transactions on Parallel and Distributed Systems
IS - 5
ER -