Comparison between electrical and free space optical interconnects for fine grain processor arrays based on interconnect density capabilities

Michael R. Feldman, Clark C. Guest, Timothy J. Drabik, Sadik Esener

Research output: Contribution to journalArticle

66 Citations (Scopus)

Abstract

Optically interconnected processor arrays are compared to conventional fully electronic processor arrays in terms of interconnect density capabilities. A complexity model is introduced that allows the calculation of the array area growth rate as an asymptotic function of the number of processing elements in the array. Lower bounds on the area growth rate of electrically interconnected processor arrays are compared to upper bounds for free-space optically interconnected circuits that employ computer generated holograms. Results indicate that for connection networks such as the hypercube, perfect shuffle and crossbar networks, that have a high minimum bisection width (a measure of the global nature of an interconnect topology) and contain some degree of spatial invariance, optically interconnected circuit area growth rates are below lower bounds on VLSI circuit growth rates.

Original languageEnglish (US)
Pages (from-to)3820-3829
Number of pages10
JournalApplied Optics
Volume28
Issue number18
DOIs
StatePublished - Sep 15 1989
Externally publishedYes

Fingerprint

free-space optical interconnects
Optical interconnects
Parallel processing systems
central processing units
Networks (circuits)
VLSI circuits
Holograms
Invariance
very large scale integration
Topology
invariance
topology
Processing
electronics

ASJC Scopus subject areas

  • Atomic and Molecular Physics, and Optics

Cite this

Comparison between electrical and free space optical interconnects for fine grain processor arrays based on interconnect density capabilities. / Feldman, Michael R.; Guest, Clark C.; Drabik, Timothy J.; Esener, Sadik.

In: Applied Optics, Vol. 28, No. 18, 15.09.1989, p. 3820-3829.

Research output: Contribution to journalArticle

@article{1a9763944b5742ef9a987dc714d2026b,
title = "Comparison between electrical and free space optical interconnects for fine grain processor arrays based on interconnect density capabilities",
abstract = "Optically interconnected processor arrays are compared to conventional fully electronic processor arrays in terms of interconnect density capabilities. A complexity model is introduced that allows the calculation of the array area growth rate as an asymptotic function of the number of processing elements in the array. Lower bounds on the area growth rate of electrically interconnected processor arrays are compared to upper bounds for free-space optically interconnected circuits that employ computer generated holograms. Results indicate that for connection networks such as the hypercube, perfect shuffle and crossbar networks, that have a high minimum bisection width (a measure of the global nature of an interconnect topology) and contain some degree of spatial invariance, optically interconnected circuit area growth rates are below lower bounds on VLSI circuit growth rates.",
author = "Feldman, {Michael R.} and Guest, {Clark C.} and Drabik, {Timothy J.} and Sadik Esener",
year = "1989",
month = "9",
day = "15",
doi = "10.1364/AO.28.003820",
language = "English (US)",
volume = "28",
pages = "3820--3829",
journal = "Applied Optics",
issn = "0003-6935",
publisher = "The Optical Society",
number = "18",

}

TY - JOUR

T1 - Comparison between electrical and free space optical interconnects for fine grain processor arrays based on interconnect density capabilities

AU - Feldman, Michael R.

AU - Guest, Clark C.

AU - Drabik, Timothy J.

AU - Esener, Sadik

PY - 1989/9/15

Y1 - 1989/9/15

N2 - Optically interconnected processor arrays are compared to conventional fully electronic processor arrays in terms of interconnect density capabilities. A complexity model is introduced that allows the calculation of the array area growth rate as an asymptotic function of the number of processing elements in the array. Lower bounds on the area growth rate of electrically interconnected processor arrays are compared to upper bounds for free-space optically interconnected circuits that employ computer generated holograms. Results indicate that for connection networks such as the hypercube, perfect shuffle and crossbar networks, that have a high minimum bisection width (a measure of the global nature of an interconnect topology) and contain some degree of spatial invariance, optically interconnected circuit area growth rates are below lower bounds on VLSI circuit growth rates.

AB - Optically interconnected processor arrays are compared to conventional fully electronic processor arrays in terms of interconnect density capabilities. A complexity model is introduced that allows the calculation of the array area growth rate as an asymptotic function of the number of processing elements in the array. Lower bounds on the area growth rate of electrically interconnected processor arrays are compared to upper bounds for free-space optically interconnected circuits that employ computer generated holograms. Results indicate that for connection networks such as the hypercube, perfect shuffle and crossbar networks, that have a high minimum bisection width (a measure of the global nature of an interconnect topology) and contain some degree of spatial invariance, optically interconnected circuit area growth rates are below lower bounds on VLSI circuit growth rates.

UR - http://www.scopus.com/inward/record.url?scp=84975645801&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84975645801&partnerID=8YFLogxK

U2 - 10.1364/AO.28.003820

DO - 10.1364/AO.28.003820

M3 - Article

AN - SCOPUS:84975645801

VL - 28

SP - 3820

EP - 3829

JO - Applied Optics

JF - Applied Optics

SN - 0003-6935

IS - 18

ER -