Optical content-addressable parallel processor: Architecture, algorithms, and design concepts

Research output: Contribution to journalArticle

25 Citations (Scopus)

Abstract

Associative processing based on content-addressable memories has been argued to be the natural solution for nonnumerical information processing applications. Unfortunately, the implementation requirements of these architectures when one uses conventional electronic technology have been cost prohibitive; therefore associative processors have not been realized. Instead, software methods that emulate the behavior of associative processing have been promoted and mapped onto conventional location- addressable systems. However, this does not bring about the natural parallelism of associative processing, namely, the ability to access many data words simultaneously. Optics has the advantage over electronics of directly supporting associative processing by providing economic and efficient interconnects, massive parallelism, and high-speed processing. The principles of designing an optical content-addressable parallel processor (OCAPP) for the efficient support of parallel symbolic computing are presented. The architecture is designed to exploit optics advantages fully in interconnects and high-speed operations. Several parallel search-and-retrieval algorithms are mapped onto an OCAPP to illustrate its capability of supporting parallel symbolic computing. A theoretical performance analysis of these algorithms is presented. This analysis reveals that the execution times of the parallel algorithms presented are independent of the problem size, which makes the OCAPP suitable for applications in which the number of data sets to be operated on is high (e.g., massive parallel processing). A preliminary optical implementation of the architecture with currently available optical components is also presented.

Original languageEnglish (US)
Pages (from-to)3241-3258
Number of pages18
JournalApplied Optics
Volume31
Issue number17
DOIs
StatePublished - Jun 10 1992

Fingerprint

Associative processing
central processing units
Optics
high speed
optics
Associative storage
associative memory
Processing
electronics
Parallel algorithms
retrieval
economics
Electronic equipment
costs
computer programs
requirements
Economics
Costs

ASJC Scopus subject areas

  • Atomic and Molecular Physics, and Optics

Cite this

Optical content-addressable parallel processor : Architecture, algorithms, and design concepts. / Louri, Ahmed.

In: Applied Optics, Vol. 31, No. 17, 10.06.1992, p. 3241-3258.

Research output: Contribution to journalArticle

@article{ce42edbb4af94c5cb5a9167ca61a771d,
title = "Optical content-addressable parallel processor: Architecture, algorithms, and design concepts",
abstract = "Associative processing based on content-addressable memories has been argued to be the natural solution for nonnumerical information processing applications. Unfortunately, the implementation requirements of these architectures when one uses conventional electronic technology have been cost prohibitive; therefore associative processors have not been realized. Instead, software methods that emulate the behavior of associative processing have been promoted and mapped onto conventional location- addressable systems. However, this does not bring about the natural parallelism of associative processing, namely, the ability to access many data words simultaneously. Optics has the advantage over electronics of directly supporting associative processing by providing economic and efficient interconnects, massive parallelism, and high-speed processing. The principles of designing an optical content-addressable parallel processor (OCAPP) for the efficient support of parallel symbolic computing are presented. The architecture is designed to exploit optics advantages fully in interconnects and high-speed operations. Several parallel search-and-retrieval algorithms are mapped onto an OCAPP to illustrate its capability of supporting parallel symbolic computing. A theoretical performance analysis of these algorithms is presented. This analysis reveals that the execution times of the parallel algorithms presented are independent of the problem size, which makes the OCAPP suitable for applications in which the number of data sets to be operated on is high (e.g., massive parallel processing). A preliminary optical implementation of the architecture with currently available optical components is also presented.",
author = "Ahmed Louri",
year = "1992",
month = "6",
day = "10",
doi = "10.1364/AO.31.003241",
language = "English (US)",
volume = "31",
pages = "3241--3258",
journal = "Applied Optics",
issn = "1559-128X",
publisher = "The Optical Society",
number = "17",

}

TY - JOUR

T1 - Optical content-addressable parallel processor

T2 - Architecture, algorithms, and design concepts

AU - Louri, Ahmed

PY - 1992/6/10

Y1 - 1992/6/10

N2 - Associative processing based on content-addressable memories has been argued to be the natural solution for nonnumerical information processing applications. Unfortunately, the implementation requirements of these architectures when one uses conventional electronic technology have been cost prohibitive; therefore associative processors have not been realized. Instead, software methods that emulate the behavior of associative processing have been promoted and mapped onto conventional location- addressable systems. However, this does not bring about the natural parallelism of associative processing, namely, the ability to access many data words simultaneously. Optics has the advantage over electronics of directly supporting associative processing by providing economic and efficient interconnects, massive parallelism, and high-speed processing. The principles of designing an optical content-addressable parallel processor (OCAPP) for the efficient support of parallel symbolic computing are presented. The architecture is designed to exploit optics advantages fully in interconnects and high-speed operations. Several parallel search-and-retrieval algorithms are mapped onto an OCAPP to illustrate its capability of supporting parallel symbolic computing. A theoretical performance analysis of these algorithms is presented. This analysis reveals that the execution times of the parallel algorithms presented are independent of the problem size, which makes the OCAPP suitable for applications in which the number of data sets to be operated on is high (e.g., massive parallel processing). A preliminary optical implementation of the architecture with currently available optical components is also presented.

AB - Associative processing based on content-addressable memories has been argued to be the natural solution for nonnumerical information processing applications. Unfortunately, the implementation requirements of these architectures when one uses conventional electronic technology have been cost prohibitive; therefore associative processors have not been realized. Instead, software methods that emulate the behavior of associative processing have been promoted and mapped onto conventional location- addressable systems. However, this does not bring about the natural parallelism of associative processing, namely, the ability to access many data words simultaneously. Optics has the advantage over electronics of directly supporting associative processing by providing economic and efficient interconnects, massive parallelism, and high-speed processing. The principles of designing an optical content-addressable parallel processor (OCAPP) for the efficient support of parallel symbolic computing are presented. The architecture is designed to exploit optics advantages fully in interconnects and high-speed operations. Several parallel search-and-retrieval algorithms are mapped onto an OCAPP to illustrate its capability of supporting parallel symbolic computing. A theoretical performance analysis of these algorithms is presented. This analysis reveals that the execution times of the parallel algorithms presented are independent of the problem size, which makes the OCAPP suitable for applications in which the number of data sets to be operated on is high (e.g., massive parallel processing). A preliminary optical implementation of the architecture with currently available optical components is also presented.

UR - http://www.scopus.com/inward/record.url?scp=84975625136&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84975625136&partnerID=8YFLogxK

U2 - 10.1364/AO.31.003241

DO - 10.1364/AO.31.003241

M3 - Article

AN - SCOPUS:84975625136

VL - 31

SP - 3241

EP - 3258

JO - Applied Optics

JF - Applied Optics

SN - 1559-128X

IS - 17

ER -