Abstract
This paper introduces a new network topology, the offset-cube, for three-dimensional image processing systems. It provides general node-to-node communications for random message traffic with an average latency that is comparable to a k-ary 3-cube. It is well suited for communication patterns common in image processing applications (e.g., image scaling, overlapping segmentation). The offset-cube topology provides an extremely compact, through-wafer implementation employing integrated optoelectronic devices. This paper describes work in progress towards the examination of this new topology and its implementation including architectural issues (e.g, oblivious and adaptive minimal routing algorithms, performance analysis with random traffic and trace-driven workloads, comparison with k-ary 3-cubes), plus implementation issues in building a scalable system.
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
L. S. Bergman, W. H. Wu, A. R. Johnston, R. Nixon, S. C. Esener, C. C. Guest, P. Yu, T. J. Drabik, M. Feldman, and S. H. Lee. Holographie Optical Interconnects for VLSI. Optical Engineering, 25:1109–1118, October 1986.
K. H. Calhoun, C. B. Camperi-Ginestet, and N. M. Jokerst. Vertical Optical Communication Through Stacked Silicon Wafers Using Hybrid Monolithic Thin Film InGaAsP Emitters and Detectors. IEEE Photonics Technology Letters, 5(2):254–257, February 1993.
R. T. Chen, H. Lu, D. Robinson, M. Wang, G. Savant, and T. Jannson. Guidedwave Planar Optical Interconnects Using Highly Multiplexed Polymer Waveguide Holograms. Journal of Lightwave Technology, 10:888–897, July 1992.
N. C. Craft and A. Y. Feldblum. Optical Interconnections Based on Arrays of Surface-emitting Lasers and Lenslets. Applied Optics, 31:1735–1739, April 1992.
J. L. Cruz-Rivera, E. V. R. DiBella, D. S. Wills, T. K. Gaylord, and E. N. Glytsis. Parallelized Implementation of the Maximum Likelihood Expectation Maximization Algorithm on a Fine-grained Optically Interconnected Architecture. to be submitted to IEEE Transactions on Medical Imaging, 1994.
William J. Dally. Performance Analysis of k-ary n-cube Interconnection Networks. IEEE Transactions on Computers, C-39(6):775–785, June 1990.
William J. Dally, Andrew Chien, Stuart Fiske, Greg Fyler, Waldemar Horwat, John Keene, Rich Lethin, Michael Noakes, Peter Nuth, and D. Scott. Wills. The Message Driven Processor: Am Integrated Multicomputer Processing Element. In ICCD '92 International Conference on Computer Design, VLSI in Computers & Processors, pages 416–419, 11–14 October 1992.
Jose Duato. A New Theory of Deadlock-Free Adapative Routing in Wormhole Networks. IEEE Transactions on Parallel and Distributed Systems, 4(12):1320–1331, December 1993.
G. C. Fox et al. Matrix Algorithms on a Hypercube I: Matrix Multiplication. Parallel Computing, 4(1), 1987.
J. M. Kallis, L. B. Duncan, S. P. Laub, M. J. Little, L. M. Miani, and D. C. Sandkulla. Reliability of the 3-D Computer Under Stress of Mechanical Vibration and Thermal Cyling. In Proceedings of the IEEE International Conference on Wafer Scale Integration, pages 65–72, 1989.
R. K. Kostuk, J. W. Goodman, and L. Hesselink. Design Considerations for Holographic Optical Interconnects. Applied Optics, 26:3947–3953, September 1987.
W. Stephen Lacy, Christophe Camperi-Ginestet, Brent Buchanan, D. Scott Wills, Nan Marie Jokerst, and Martin Brooke. A Fine-Grain, High-Throughput Architecture Using Through-Wafer Optical Interconnect. In Proceedings of First Annual Workshop on Massively Parallel Processing and Optical Interconnect, pages 27–36, 26–29 April 1994.
W. Stephen Lacy, Jose Cruz-Rivera, and D. Scott Wills. A Scalable Optical Interconnection Network for Fine-Grain Parallel Architectures. to be submitted to IEEE Transactions on Computers, October 1994.
M. J. Little, R. D. Etchells, J. Grinberg, S. P. Laub, J. G. Nash, and M. W. Yung. The 3-D Computer. In Proceedings of the IEEE International Conference on Wafer Scale Integration, pages 55–64, 1989.
Y. S. Liu and J. Bristow. Hybrid Integration of Electrical and Optical Interconnect for MCMs Applications. In Proceeding of the IEEE Lasers and Electro-Optics Summer Topical Meeting on Hybrid Optoelectric Integration and Packaging, page 10, July 1993.
Michael Noakes and William J. Dally, System Design of the J-Machine. In Sixth MIT Conference of Advanced Research in VLSI, pages 179–194. MIT Press, 1990.
Daniel A. Reed and Richard M. Fujimoto. Multicomputer Networks: Message-Based Parallel Processing. Scientific Computation Series. MIT Press, 1987.
Gregory K. Wallace. Overview of the JPEG (ISO/CCITT) Still Image Compression Standard. In Richard Feinberg, editor, Current Overviews in Optical Science and Engineering I, pages 358–371. SPIE Optical Engineering Press, Bellingham, Washington, 1990.
D. Scott Wills, Huy Cat, Jose Cruz-Rivera, W. Stephen Lacy, James M. Baker, John Eble, Abelardo Lopez-Lagunas, and Mike Hopper. High Throughput, Low Memory Applications on the Pica Architecture. submitted to IEEE Transactions on Parallel and Distributed Systems, May 1994.
D. Scott Wills and Matthias Grossglauser. A Scalable Optical Interconnection Network for Fine-Grain Parallel Architectures. In 1993 International Conference on Parallel Processing, pages I-154–I-157, 16–20 August 1993.
D. Scott Wills and Matthias Grossglauser. A Three-Dimensional Optical Interconnection Network for Fine-Grain Parallel Architectures. In Proceeding of the IEEE Lasers and Electro-Optics Summer Topical Meeting on Hybrid Optoelectric Integration and Packaging, pages 21–22, 26–28 July 1993.
D. Scott Wills, W. Stephen Lacy, Huy Cat, Michael A. Hopper, Ashutosh Razdan, and Sek M. Chai. Pica: An Ultra-Light Processor for High-Throughput Applications. In ICCD '93 International Conference on Computer Design, VLSI in Computers & Processors, pages 410–414, 3–6 October 1993.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1994 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Wills, D.S., Lacy, W.S., Cruz-Rivera, J. (1994). The offset cube: An optoelectronic interconnection network. In: Bolding, K., Snyder, L. (eds) Parallel Computer Routing and Communication. PCRCW 1994. Lecture Notes in Computer Science, vol 853. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-58429-3_30
Download citation
DOI: https://doi.org/10.1007/3-540-58429-3_30
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-58429-2
Online ISBN: 978-3-540-48787-6
eBook Packages: Springer Book Archive