Reducing communication latency with path multiplexing: in optically interconnected multiprocessor systems

A physical link can be time-multiplexed to create several time slots, each of which corresponding to a virtual link. A conventional approach establishes a connection along a path using a set of independent time slots (or virtual links) and thus requires the use of switching devices capable of interchanging time slots. This paper proposes a different approach to all-optical Time Division Multiplexed (TDM) communications in multiprocessor systems. The idea is to establish a connection along a path using a set of time slots (or virtual links) that are dependent on each other, so that no time-slot interchanging is required. It is found that, despite of the possibility that establishing a connection may take a longer time, the proposed approach will result in lower overall communication latency as it eliminates the delays introduced by the time-slot interchanging switching devices.<<ETX>>

[1]  F. Leighton,et al.  Introduction to Parallel Algorithms and Architectures: Arrays, Trees, Hypercubes , 1991 .

[2]  P. Messina,et al.  Architectural requirements of parallel scientific applications with explicit communication , 1993, ISCA '93.

[3]  Rami G. Melhem,et al.  Pipelined Communications in Optically Interconnected Arrays , 1991, J. Parallel Distributed Comput..

[4]  F.J. Leonberger,et al.  Optical interconnections for VLSI systems , 1984, Proceedings of the IEEE.

[5]  C. R. Jesshope,et al.  High performance communications in processor networks , 1989, ISCA '89.

[6]  David K. Hunter,et al.  New architectures for optical TDM switching , 1993 .

[7]  Chris R. Jesshope,et al.  High Performance Communications In Processor Networks , 1989, The 16th Annual International Symposium on Computer Architecture.

[8]  Rami G. Melhem,et al.  Dynamic Reconfiguration of Optically Interconnected Networks with Time-Division Multiplexing , 1994, J. Parallel Distributed Comput..

[9]  Chuan-lin Wu,et al.  Performance Analysis of Multistage Interconnection Network Configurations and Operations , 1992, IEEE Trans. Computers.

[10]  Rami G. Melhem,et al.  Space Multiplexing of Waveguides in Optically Interconnected Multiprocessor Systems , 1989, Comput. J..

[11]  Lionel M. Ni,et al.  A conflict-free memory design for multiprocessors , 1991, Proceedings of the 1991 ACM/IEEE Conference on Supercomputing (Supercomputing '91).

[12]  Aura Ganz,et al.  A time-wavelength assignment algorithm for a WDM star network , 1992, [Proceedings] IEEE INFOCOM '92: The Conference on Computer Communications.

[13]  Tse-yun Feng,et al.  A Survey of Interconnection Networks , 1981, Computer.

[14]  S H Lee,et al.  Comparison between optical and electrical interconnects based on power and speed considerations. , 1988, Applied optics.

[15]  Rami Melhem,et al.  Array processors with pipelined optical busses , 1990, [1990 Proceedings] The Third Symposium on the Frontiers of Massively Parallel Computation.

[16]  Bryan E. Floren.  Optical interconnects in the Touchstone supercomputer program , 1992 .

[17]  Kavita Bala,et al.  Algorithms for routing in a linear lightwave network , 1991, IEEE INFCOM '91. The conference on Computer Communications. Tenth Annual Joint Comference of the IEEE Computer and Communications Societies Proceedings.

[18]  Harry F. Jordan,et al.  Serial array shuffle-exchange architecture for universal permutation of time-slots , 1990, Photonics West - Lasers and Applications in Science and Engineering.

[19]  Imrich Chlamtac,et al.  Lightpath communications: an approach to high bandwidth optical WAN's , 1992, IEEE Trans. Commun..

[20]  Richard B. Gillett.  Memory Channel Network for PCI , 1996, IEEE Micro.

[21]  Thomas A. Lane,et al.  Optical interconnects for interprocessor communications in the Connection Machine , 1989, Proceedings 1989 IEEE International Conference on Computer Design: VLSI in Computers and Processors.

[22]  Aura Ganz,et al.  Optical interconnects for multiprocessors cost performance trade-offs , 1992, [Proceedings 1992] The Fourth Symposium on the Frontiers of Massively Parallel Computation.

[23]  S. F. Nugent,et al.  The iPSC/2 direct-connect communications technology , 1988, C3P.

[24]  William J. Dally,et al.  Deadlock-Free Message Routing in Multiprocessor Interconnection Networks , 1987, IEEE Transactions on Computers.

[25]  Rami G. Melhem,et al.  Time-Division Optical Communications in Multiprocessor Arrays , 1993, IEEE Trans. Computers.

[26]  Richard A. Thompson.  Architectures with improved signal-to-noise ratio in photonic systems with fiber-loop delay lines , 1988, IEEE J. Sel. Areas Commun..

[27]  V. Li,et al.  A Wavelength-Convertible Optical Network , 1993 .

[28]  Lionel M. Ni,et al.  A survey of wormhole routing techniques in direct networks , 1993, Computer.

[29]  Rami G. Melhem,et al.  Reconfiguration with Time Division Multiplexed MIN's for Multiprocessor , 1994, IEEE Trans. Parallel Distributed Syst..

[30]  Kumar N. Sivarajan,et al.  Optimal routing and wavelength assignment in all-optical networks , 1994, Proceedings of INFOCOM '94 Conference on Computer Communications.

[31]  Richard M. Fujimoto,et al.  Multicomputer Networks: Message-Based Parallel Processing , 1987 .

[32]  Harry F. Jordan,et al.  Time multiplexed optical computers , 1991, Proceedings of the 1991 ACM/IEEE Conference on Supercomputing (Supercomputing '91).

[33]  F. Cappello,et al.  Toward high communication performance through compiled communications on a circuit switched interconnection network , 1995, Proceedings of 1995 1st IEEE Symposium on High Performance Computer Architecture.

[34]  C. Qiao,et al.  Reconfiguration With Time Division Multiplexed MINs for Multiprocessor Communications , 1994 .

[35]  Kalyani Bogineni,et al.  Hierarchical Scalable Photonic Architectures for High-Performance Processor Interconnection , 1993, IEEE Trans. Computers.