Parallel Computing with “Intelligent Optical Networks”

  • Ted H. Szymanski
Part of the The Springer International Series in Engineering and Computer Science book series (SECS, volume 468)


Parallel computing on a class of networks called “Intelligent Optical Networks” is examined. Traditional All-Optical and Passive-Optical networks have very limited abilities to process data in the all-optical layer. In contrast, we propose the concept of an “Intelligent Optical Network”, which has the ability to process and buffer vast amounts of data within the optoelectronic layer. Intelligent Optical Networks are can be realized with CMOS substrates with optical I/O which have been flip-chip bonded over the surface. This technology has the potential for thousands of optical I/O per substrate, and supports rapid and transparent conversion of parallel bits between the electrical and optical domains. The design of an Intelligent Optical Network for a Network-of-Workstations (NOW) is described. Each workstation has a parallel fiber ribbon optical datalink to a centralized switching core, implemented on a single optoelectronic CMOS integrated circuit. A centralized integrated circuit inter-connecting 32 fiber ribbons, each with 32 fibers clocked at 1 GHz, would support an aggregate bandwidth of 1 Tb/s. The switching is based on a conventional “Broadcast-and-Select” architecture. We consider the impact of an Intelligent Optical Network on the parallel computing of Butterfly-based algorithms. It is shown that a multi-dimensional NOW with DSP coprocessors can sustain performances in the 100 TeraFLOP per second range for Butterfly-based scientific algorithms. This class of interconnects thus represents one viable approach to achieve the US Accelerated Strategic Computing Initiative computing targets of 100 TeraFLOP machines by the year 2004.


Shared Memory Clock Cycle Digital Signal Processor Optical Interconnection Vertical Cavity Surface Emit Laser 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    T.E. Anderson, D.E. Culler, and D.A. Patterson, “A Case for NOW (Networks of Workstations)”, IEEE Micro, vol. 16,no. 1, Feb. 1995, pp. 54–64.CrossRefGoogle Scholar
  2. [2]
    T.E. Anderson, S.S. Owicki, J.B. Saxe, and C.P. Thacker, “High-Speed Switch Scheduling for Local-Area networks”, ACM Transactions on Computer Systems, vol. 11,no. 4, Nov. 1993, pp. 319–352.CrossRefGoogle Scholar
  3. [3]
    A. Benner, L. Rudolph, E. Schenfeld, T. Sterling, T. Szymanski, “Panel Discussion: Design Options for Interconnecting 100+ TFLOP/sec Parallel Supercomputer in 2004”, IEEE 4th Int. Conf. Massively Parallel Processing using Optical Interconnects (MPPOI), June 1998.Google Scholar
  4. [4]
    H. Bourdin, A. Ferreire and K. Marcus, “A Comparative Study of One-to-Many WDM Lightwave Interconnection Networks for Multiprocessors”, 2nd Int. Conf. Massively Parallel Processing using Optical Interconnects (MPPOI), Oct. 1995, pp. 257–264.Google Scholar
  5. [5]
    D. Chiarulli, S. Levitan, R. Melhem, J. Teza, and G. Gravenstreter, “Partitioned Optical Passive Star (POPS) Topologies Multiprocessor Interconnection Networks with Distributed Control”, IEEE Journal of Lightwave Technology, vol. 14,no. 7, 1996, pp. 1601–1612.CrossRefGoogle Scholar
  6. [6]
    A. Chien, “A Cost and Speed Model for k-ary n-Cube Wormhole Routers”, IEEE Transactions on Parallel and Distributed Systems, vol. 9,no. 2, Feb. 1998.Google Scholar
  7. [7]
    D. Clark, “Breaking the Teraflops Barrier”, IEEE Computer, vol. 30, Feb. 1997, pp. 12–14.Google Scholar
  8. [8]
    W.A. Crossland and T.D. Wilkinson, “Optically Transparent Switching in Telecommunications Using Ferroelectric Liquid Crystals Over Silicon VLSI Circuits”, Proceedings of the 1996 IEEE/LEOS Summer Topical Meeting, 1996, pp. 22–23.Google Scholar
  9. [9]
    D.R. Engebretsen, D. M. Kuchta, R. C. Booth, J.D. Crow, and W.G. Nation, “Parallel Fiber-Optic SCI Links,” IEEE Micro, vol. 16,no. 1, Feb. 1996, pp. 20–26.CrossRefGoogle Scholar
  10. [10]
    M. Galles, “Spider: A High-Speed Network Interconnect”, IEEE Micro, vol. 17,no. 1, Jan/Feb. 1997, pp. 34–39.CrossRefMathSciNetGoogle Scholar
  11. [11]
    A.V. Gerbessiotis and L. Valiant, “Direct Bulk-synchronous parallel algorithm”, Journal of Parallel and Distributed Computing, vol. 22, 1994, pp. 251–257.CrossRefGoogle Scholar
  12. [12]
    Glesk, I. Kang, K.I. Prucnal, P. R., “Demonstration of ultrafast all-optical packet routing”, Electronics Letters, vol. 33,no. 9, Apr 1997, pp. 794–795.CrossRefGoogle Scholar
  13. [13]
    J.L. Hennessy and D.A. Patterson, Computer Architecture: A Quantitative Approach, Second Edition, Morgan Kauffman, 1995.Google Scholar
  14. [14]
    S.J. Hinterlong, A.L. Lentine, D.J. Reiley, J.M. Sasian, R.L. Morrison, R.A. Novotny, M.G. Beckman, D.B. Buchholz, T.J. Cloonan, and G.W. Richards, “An ATM Switching System Demonstration Using a 40 Gb/s Throughput Smart Pixel Opto-electronic VLSI Chip”, Proc. 1996 IEEE/LEOS Summer Topical Meeting, 1996, pp. 47–48.Google Scholar
  15. [15]
    D. Knuth, The Art of Computer Programming, Vol. 3, Sorting and Searching, Second Edition, Addison-Wesley, 1973.Google Scholar
  16. [16]
    A.V. Krishnamoorthy, and D.A.B. Miller, “Scaling Optoelectronic-VLSI Circuits into the 21st Century: A Technology Roadmap”, IEEE Journal of Selected Topics in Quantum Electronics, vol. 2,no. 1, Apr. 1996, pp. 55–76.CrossRefGoogle Scholar
  17. [17]
    A.V. Krishnamoorthy, J.E. Ford, K.W. Goossen, J.A. Walker, B. Tseng, S.P. Hui, J.E. Cunningham, W. Y. Jan, T. K. Woodward, M.C. Nuss, R. G. Rozier, F. E. Kiamilev, and D. A. B. Miller, “The AMOEBA Chip: An Optoelectronic Switch for Multiprocessor Networking Using Dense-WDM”, Proceedings of the Third Int. Conf. on Massively Parallel Processing Using Optical Interconnections (MPPOI’96), 1996, pp. 94–100Google Scholar
  18. [18]
    A.L. Lentine, K.W. Goossen, J.A. Walker, L.M.F. Chirovsky, L.A. DAsaro, S.P. Hui, B.T. Tseng, R.E. Leibenguth, J.E. Cunningham, W.Y. Jan, J.M. Kuo, D. Dahringer, D. Kossives, D.D. Bacon, G. Livescu, R.L. Morrison, R.A. Novotny, and D.B. Buchholz, “Optoelectronic VLSI Switching Chip with Greater Than 4,000 Optical Flip-chip-bonded GaAs/AlGaAS MQW Modulators and Detectors on Silicon CMOS Circuitry”, Conf. Lasers and Electrooptics, 1996, pp. 517–518.Google Scholar
  19. [19]
    S. Matsuo, T. Nakahara, Y. Kohama, Y. Ohiso, S. Fukushima, and T. Kurokawa, “Monolithically Integrated Photonic Switching Device Using an MSM PD, MESFET’s, and a VCSEL”, IEEE Photonics Technology Letters, vol. 7,no. 10, Oct. 1995, pp. 1165–1167.CrossRefGoogle Scholar
  20. [20]
    Semiconductor Industry Association, “The National Technology Roadmap for Semiconductors”, SIA, San Jose, 1997.Google Scholar
  21. [21]
    Shell, M. Vaughn, M D. Wang, A. Blumenthal, D J. Rigole, P-J. Nilsson, S., “Experimental demonstration of an all-optical routing node for multihop wavelength routed networks”, IEEE Photonics Technology Letters, vol. 8,no. 10 Oct 1996, p 1391–1393.CrossRefGoogle Scholar
  22. [22]
    S. Sherif, T.H. Szymanski, and H.S. Hinton, “Design and Implementation of a Field Programmable Smart Pixel Array”, Proc. 1996 IEEE/LEOS Summer Topical Meeting, 1996.Google Scholar
  23. [23]
    B. Supmonchai and T.H. Szymanski, “High Speed VLSI Concentrators for Terabit Intelligent Optical Backplanes”, Optics in Computing 1998, June, 1998.Google Scholar
  24. [24]
    T.H. Szymanski and H.S. Hinton, “Reconfigurable Intelligent Optical Backplane for Parallel Computing and Communications”, Applied Optics, vol. 35,no. 8, Mar. 1996, pp. 1253–1268.CrossRefGoogle Scholar
  25. [25]
    T.H. Szymanski and H.S. Hinton, “Architecture of a Terabit Free-Space Intelligent Optical Backplane”, Journal of Parallel and Distributed Computing, to appear, 1998/99.Google Scholar
  26. [26]
    T.H. Szymanski, A. Au, M. Lafrenire-Roula, V. Tyan, B. Supmonchai, J. Wong, B. Zerrouk, and S.T. Obenaus, “Terabit Optical Local Area Networks for Multiprocessing Systems”, Applied Optics, Special Issue on Massively Parallel Optical Interconnects for Multiprocessor Systems, vol. 37,no. 2, Jan 1998, pp.264–275.Google Scholar
  27. [27]
    T.H. Szymanski, “Design Principles for Practical Self-Routing Nonblocking Switching Networks with O(N log N) Bit Complexity”, IEEE Transactions on Computers, vol. 46,no. 10, Oct. 1997, pp. 1057–1069.CrossRefGoogle Scholar
  28. [28]
    T.H. Szymanski, “Hypermeshes — Optical Interconnection Networks for Parallel Computing”, Journal of Parallel and Distributed Computing, vol. 26, Apr. 1995, pp. 1–23.MATHCrossRefGoogle Scholar
  29. [29]
    T.H. Szymanski and B. Supmonchai, “Reconfigurable computing with Optical Interconnects — an Economic Argument for Optics”, Proceedings of the Third Int. Conf. on Massively Parallel Processing Using Optical Interconnections (MPPOI’96), 1996.Google Scholar
  30. [30]
    S.K. Tweksbury, K. Devabattini, and V. Gandikota “A Parallel DSP testbed with a Heterogeneous and Reconfigurable Network Fabric”, Proceedings of the Int. Conf. Innovative Systems on Silicon, ISIS 1997, 1997, pp. 310–322.Google Scholar

Copyright information

© Kluwer Academic Publishers 1998

Authors and Affiliations

  • Ted H. Szymanski
    • 1
  1. 1.Department of Electrical EngineeringMcGill UniversityMontrealCanada

Personalised recommendations