Packet Synchronization for Synchronous Optical Deflection-Routed Interconnection Networks

Deflection routing resolves output port contention in packet switched multiprocessor interconnection networks by granting the preferred port to the highest priority packet and directing contending packets out other ports. When combined with optical links and switches, deflection routing yields simple bufferless nodes, high bit rates, scalable throughput, and low latency. We discuss the problem of packet synchronization in synchronous optical deflection networks with nodes distributed across boards, racks, and cabinets. Synchronous operation is feasible due to very predictable optical propagation delays. A routing control processor at each node examines arriving packets and assigns them to output ports. Packets arriving on different input ports must be bit wise aligned; there are no elastic buffers to correct for mismatched arrivals. "Time of flight" packet synchronization is done by balancing link delays during network design. Using a directed graph network model, we formulate a constrained minimization problem for minimizing link delays subject to synchronization and packaging constraints. We demonstrate our method on a ShuffleNet graph, and show modifications to handle multiple packet sizes and latency critical paths.

[1]  Frank M. Pittelli,et al.  Analysis of a 3D toroidal network for a shared memory architecture , 1988, Proceedings. SUPERCOMPUTING '88.

[2]  Anthony S. Acampora,et al.  Multihop lightwave networks: a comparison of store-and-forward and hot-potato routing , 1992, IEEE Trans. Commun..

[3]  P. Baran,et al.  On Distributed Communications Networks , 1964 .

[4]  G. Zorpette,et al.  Supercomputers-the power of parallelism , 1992 .

[5]  Cecilia R. Aragon,et al.  Optimization by Simulated Annealing: An Experimental Evaluation; Part I, Graph Partitioning , 1989, Oper. Res..

[6]  D.J. Blumenthal,et al.  First demonstration of multihop all-optical packet switching , 1994, IEEE Photonics Technology Letters.

[7]  Bruce E. Hajek,et al.  Performance of shuffle-like switching networks with deflection , 1990, Proceedings. IEEE INFOCOM '90: Ninth Annual Joint Conference of the IEEE Computer and Communications Societies@m_The Multiple Facets of Integration.

[8]  Charles E. Leiserson,et al.  Optimizing synchronous systems , 1981, 22nd Annual Symposium on Foundations of Computer Science (sfcs 1981).

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

[10]  W. Daniel Hillis,et al.  The CM-5 Connection Machine: a scalable supercomputer , 1993, CACM.

[11]  Ragai L. Khalil.  Clock Skew Analysis For Si And GaAs Receivers In Optical Clock Distribution Systems , 1990, Other Conferences.

[12]  I. Malitson.  Interspecimen Comparison of the Refractive Index of Fused Silica , 1965 .

[13]  H. T. Kung,et al.  Synchronizing Large VLSI Processor Arrays , 1985, IEEE Trans. Computers.

[14]  Donald Yeung,et al.  THE MIT ALEWIFE MACHINE: A LARGE-SCALE DISTRIBUTED-MEMORY MULTIPROCESSOR , 1991 .

[15]  Norman Balabanian,et al.  Electrical Network Theory , 1969 .

[16]  A. Gibbons.  Algorithmic Graph Theory , 1985 .

[17]  David Callahan,et al.  A future-based parallel language for a general-purpose highly-parallel computer , 1990 .

[18]  Davis H. Hartman,et al.  Digital High Speed Interconnects: A Study Of The Optical Alternative , 1986 .

[19]  H.T. Mouftah,et al.  Photonic fast packet switching , 1993, IEEE Communications Magazine.

[20]  Harry F. Jordan,et al.  Propagation delay uncertainty in time-of-flight systems , 1995 .

[21]  M. J. Karol,et al.  ShuffleNet: an application of generalized perfect shuffles to multihop lightwave networks , 1988, IEEE INFOCOM '88,Seventh Annual Joint Conference of the IEEE Computer and Communcations Societies. Networks: Evolution or Revolution?.

[22]  D. H. Hartman,et al.  Optical clock distribution using a mode-locked semiconductor laser diode system , 1991 .

[23]  Harry F. Jordan,et al.  An extended fiber-optic backplane for multiprocessors , 1994, 1994 Proceedings of the Twenty-Seventh Hawaii International Conference on System Sciences.

[24]  Albert G. Greenberg,et al.  Deflection routing in hypercube networks , 1992, IEEE Trans. Commun..

[25]  Anoop Gupta,et al.  The Stanford Dash multiprocessor , 1992, Computer.

[26]  Kenneth E. Batcher.  Decomposition of Perfect Shuffle Networks , 1991, ICPP.

[27]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[28]  Jon R. Sauer,et al.  Design and implementation of a prototype optical deflection network , 1994, SIGCOMM 1994.

[29]  John E. Midwinter,et al.  Photonics in switching: the next 25 years of optical communications? , 1992 .

[30]  R J Feuerstein,et al.  Implementation of a general-purpose stored-program digital optical computer. , 1994, Applied optics.

[31]  Harry F. Jordan,et al.  Optoelectronic time-of-flight design and the demonstration of an all-optical, stored program, digital computer , 1994 .

[32]  Erik Hagersten,et al.  DDM - A Cache-Only Memory Architecture , 1992, Computer.

[33]  Flaminio Borgonovo,et al.  Circuit service in deflection networks , 1991, IEEE INFCOM '91. The conference on Computer Communications. Tenth Annual Joint Comference of the IEEE Computer and Communications Societies Proceedings.

[34]  Flaminio Borgonovo,et al.  A Versatile Model for Predicting the Performance of Deflection-Routing Networks , 1992, Perform. Evaluation.