Massively Parallel Computers Using Optical Interconnects — The Synoptique Project —

  • P. Scheer
  • T. Collette
  • P. Churoux
Chapter

Abstract

The use of computers in industry, research centres, telecommunications and at home is growing very quickly. When the applications are time consuming, such as image or signal processing, embedded processing or CAD software, the use of powerful parallel computers becomes a necessity. For this reason, we began to design, in 1987, a parallel computer, named Sympati2, dedicated to low level image processing. Six years later, we designed another parallel computer, called Symphonie, which is well suited for iconic and symbolic data processing. We are now working on a new generation of parallel computers which will use fast electronic ASIC based on 0.25(m standard cell technology. However, although the clock frequency can be very high in those components, this is not the case for the speed of the inputs and outputs. In order to fully take advantage of powerful components within parallel architectures, we started two years ago a research project named Synoptique in which we are studying the advantages of optical interconnects in our domain. This project is carried out in collaboration with the Optical Department of ONERA CERT.

Keywords

Expense Eter PCBs Paral 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    P. J. Ayliffe, J. W. Parker, and A. Robinson. Comparison of optical and electrical data interconnections at the board and backplane levels. In P. SPIE, editor, Optical Interconnections and Networks, volume 1281, pages 2 - 15, 1990.Google Scholar
  2. [2]
    R. F. Carson, M. L. Lovejoy, K. L. Lear, M. E. Warren, P. K. Seigal, G. A. Patrizi, S. P. Kilcoyne, and D. C. Craft. Low-power modular parallel photonic data links. In 46th ECTC Conference, pages 321 - 326, Orlando, May 1996.Google Scholar
  3. [3]
    P. Churoux, J.-P. Bouzinac, S. K. M. Fracès, D. Comte, N. Hifdi, T. Collette, and P. Scheer. Optical Interconnections for Parallel Processing, chapter Optical free-space interconnections inside parallel architectures: ONERA-CERT activities. Kluwer Academic Publishers, 1997. P. Berthomé and A. Ferreira, editors (This Issue).Google Scholar
  4. [4]
    T. Collette, H. Essafi, D. Juvin, and J. Kaiser. Sympati X: a SIMD computer performing the low and intermediate levels of image processing. Future Generation Computer System, 10 (1), Apr. 1994.Google Scholar
  5. [5]
    T. Collette, C. Gamrat, D. Juvin, J.-F. Larue, L. Letellier, R. Schmit, and M. Viala. SYMPHONIE calculateur massivement parallèle: modélisation et réalisation. In Journées Adéquation Algorithme Architecture en Traitement du signal et images, Toulouse, Jan. 1996. in French.Google Scholar
  6. [6]
    M. R. Feldman, S. C. Esener, C. C. Guest, and S. H. Lee. Comparison between optical and electrical interconnects based on power and speed considerations. Applied Optics, 27 (9): 1742 - 1751, 1988.ADSCrossRefGoogle Scholar
  7. [7]
    J. W. Goodman, F. I. Leonberger, S.-Y. Kung, and R. A. Athale. Optical interconnections for VLSI systems. Proceedings IEEE, 72 (7): 850 - 865, 1984.ADSCrossRefGoogle Scholar
  8. [8]
    A. Hussain, T. Lane, C. Sullivan, J. Bristow, and A. Guha. Optical backplanes for massively parallel processors and demonstration in the Connection Machine. In GOMAC, Las Vegas, Nov. 1990.Google Scholar
  9. [9]
    D. Juvin, J.-L. Basille, H. Essafi, and J.-Y. Latil. SYMPATI2, a 1.5 D processor array for image application. In EUSIPCO, Grenoble, 1988.Google Scholar
  10. [10] S. Koh, L. Ye, H. W. Carter, and J. T. Boyd. Optoelectronic interconnect simulation using a mixed mode simulator. In SPIE, editor, Optoelectronic Interconnects III,volume 2400, pages 244-251, San Jose, Feb. 1995.Google Scholar
  11. [11]
    B. S. Landman and R. L. Russo. On a pin versus block relationship for partitions of logic graphs. IEEE Transactions on Computers, C-21: 1469 - 1479, Feb. 1971.Google Scholar
  12. [12]
    S. P. Levitan, P. J. Marchand, M. A. Rempel, D. M. Chiarulli, and F. B. McCormick. Computer-aided design if free-space optoelectronic interconnection (FSOI) systems. In 2nd International Conference on Massively Parallel Processing Using Optical Interconnections (MPPOI'95), pages 239 - 245, San Antonio, Oct. 1995.CrossRefGoogle Scholar
  13. [13]
    T. Maurin and F. Devos. Optical approaches to overcome present limitations for intercommunication and control in parallel electronic architectures. In P. SPIE, editor, Optics for computers: architectures and technologies, volume 1505, pages 158 - 165, 1991.Google Scholar
  14. [14]
    L. L. Moresco. Electronic system packaging: the search for manufacturing the optimum in a sea of constraints. IEEE Transactions on Components, Packaging, and Manufacturing Technology, 13 (3): 494 - 508, 1990.Google Scholar
  15. [15]
    K. Obermann, S. Kindt, and K. Petermann. Turn-on jitter in zero-biased single-mode semicondutor lasers. IEEE Photonics Technology Letters, 8 (1): 31 - 33, Jan. 1996.ADSCrossRefGoogle Scholar
  16. [16]
    J. W. Parker. Optical interconnection for advanced processor systems: a review of the ESPRIT II OLIVES program. Journal of Lightwave Technology, 9 (12): 1764 - 1773, 1991.ADSCrossRefGoogle Scholar
  17. [17]
    K. Preston. The Abingdon cross benchmark survey. IEEE Computer, July 1989.Google Scholar
  18. [18]
    T. Sakano, T. Matsumoto, and K. Noguchi. Three-dimensional board-to-board free-space optical interconnects and their application to the prototype multiprocessor system: COSINE-III. Applied Optics, 34(11):1815-1822, 1995.Google Scholar
  19. [19]
    P. Scheer, T. Collette, D. Juvin, A. Chenevas-Paule, J.-P. Bouzinac, P. Churoux, and M. Fracès. A massively parallel SIMD multi-processor system using optical interconnects: SYNOPTIQUE. In Optical Computing Conference (OC'96), pages 120-121,Sendai, Japan, Apr. 1996.Google Scholar
  20. [20]
    J.-M. Wang, E. Kanterakis, A. Katz, Y. Zhang, Y. Li, and N. Murray. High-speed free-space interconnect based on optical ring topology: experimental demonstration. Applied Optics, 33(26):6181-6187, 1994.Google Scholar
  21. [21]
    C. Weems. The next generation image understanding architecture. In ARPA IUW,pages 1133-1140,Monterey, CA, 1994.Google Scholar
  22. [22]
    D. S. Wills, W. S. Lacy, C. Camperi-Ginestet, B. Buchanan, H. H. Cat, S. Wilkinson, M. Lee, N. M. Jokerst, and M. A. Brooke. A three-dimensional high-throughput architecture using though-wafer optical interconnect. Journal of Lightwave Technology, 13(6):1085-1092, 1995.Google Scholar

Copyright information

© Springer Science+Business Media Dordrecht 1998

Authors and Affiliations

  • P. Scheer
    • 1
  • T. Collette
    • 1
  • P. Churoux
    • 2
  1. 1.LETI (CEA-Technologies Avancées)DEIN - CEA SaclayGif-Sur-YvetteFrance
  2. 2.Cert-OneraToulouseFrance

Personalised recommendations