An Embedded Optical Vector Matrix Multiplication Processor

Hui Kai Fu

doi:10.4028/www.scientific.net/AMM.263-266.1334

Paper Titles

A Parallel Block Predictor-Corrector Method by Python-Based Distributed Computing
p.1315

Study on Intelligent Monitoring and Controlling System in Oil Drilling Optimization
p.1319

The Simulation Study on Environmental Field of Livestock and Poultry Establishments under Two Winds Conditions
p.1323

Research on Dynamic Micro-Extended-Analog-Computer Array Controller
p.1327

An Embedded Optical Vector Matrix Multiplication Processor
p.1334

A Projection-Based Locality-Sensitive Hashing Technique for Reducing False Negatives
p.1341

A Semantic-Oriented Method for Ontology Merging
p.1347

Application of Visual Basic and FORTRAN Mixed-Language Programming to Visualization Technology for Hydro-Meteorological Model
p.1352

Campus Network Upgrade and Optimization Based on OPNET
p.1356

HomeApplied Mechanics and MaterialsApplied Mechanics and Materials Vols. 263-266An Embedded Optical Vector Matrix Multiplication...

An Embedded Optical Vector Matrix Multiplication Processor

Abstract:

In this paper, an embedded architecture of optical vector matrix multiplier (OVMM) is presented. The embedded architecture is aimed at optimizing the data flow of vector matrix multiplier (VMM) to promote its performance. The performance according to the architecture is analyzed and the simulation shows that Amdahl's law is used to analyze the hybrid opto-electronic system, and the electronic part and its interaction with optical part form the bottleneck of system.

You might also be interested in these eBooks

View Preview

Info:

Periodical:

Applied Mechanics and Materials (Volumes 263-266)

Pages:

1334-1337

DOI:

https://doi.org/10.4028/www.scientific.net/AMM.263-266.1334

Citation:

Cite this paper

Online since:

December 2012

Authors:

Hui Kai Fu

Keywords:

Cluster System, OVMM, VMM

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] Goodman, J. W., Dias, A. R., and Woody, L. M.: 'Fully parallel, high-speed incoherent optical method for performing discrete Fourier transforms', Opt. Lett., 1978, 2, (1), p.1.

DOI: 10.1364/ol.2.000001

Google Scholar

[2] Casasent, D. and Jackson, J.: 'Space and frequency-multiplexed optical linear algebra processor: fabrication and initial tests', Appl. Opt., 1986, 25, (14), pp.2258-2263.

DOI: 10.1364/ao.25.002258

Google Scholar

[3] Sawchuk, A. A., Jenkins, B. K., Raghavendra, C. S., and Varma, A.: 'Optical crossbar networks', Computer, 1987, 20, (6), pp.50-60.

DOI: 10.1109/mc.1987.1663591

Google Scholar

[4] Carter III, J. A., Pape, D. R., Wasilousky, P. A., and Sunderlin, T. A.: 'High-performance optical vector-matrix coprocessor'. Photonics for Processors, Neural Networks, and Memories II, San Diego, CA, USA, 1994, pp.225-236.

DOI: 10.1117/12.187290

Google Scholar

[5] Mosca, E. P., Griffin, R. D., Pursel, F. P., and Lee, J. N.: 'Acoustooptical matrix-vector product processor: implementation issues', Appl. Opt., 1989, 28, (18), pp.3843-3851.

DOI: 10.1364/ao.28.003843

Google Scholar

[6] Gruber, M., Jahns, J., and Sinzinger, S.: 'Planar-integrated optical vector-matrix multiplier', Appl. Opt., 2000, 39, (29), pp.5367-5373.

DOI: 10.1364/ao.39.005367

Google Scholar

[7] Verber, C. M.: 'Integrated-optical approaches to numerical optical processing', Proceedings of the IEEE, 1984, 72, (7), pp.942-953.

DOI: 10.1109/proc.1984.12950

Google Scholar

[8] Flynn, M. J.: 'Very high-speed computing systems'. Proceedings of the IEEE, 1966, pp.1901-1909.

Google Scholar