A neural-network packet switch controller: Scalability, performance, and network optimization

Keith J. Symington; Andrew J. Waddie; Mohammad R. Taghizadeh; John F. Snowdon

doi:10.1109/TNN.2002.806646

A neural-network packet switch controller: Scalability, performance, and network optimization

Keith J. Symington, Andrew J. Waddie, Mohammad R. Taghizadeh, John F. Snowdon

School of Engineering & Physical Sciences

Research output: Contribution to journal › Article › peer-review

12 Citations (Scopus)

Abstract

This paper examines a novel combination of architecture and algorithm for a packet switch controller that incorporates an experimentally implemented optically interconnected neural network. The network performs scheduling decisions based on incoming packet requests and priorities. We show how and why, by means of simulation, the move from a continuous to a discrete algorithm has improved both network performance and scalability. The system's limitations are examined and conclusions drawn as to its maximum scalability and throughput based on today's technologies.

Original language	English
Pages (from-to)	28-34
Number of pages	7
Journal	IEEE Transactions on Neural Networks
Volume	14
Issue number	1
DOIs	https://doi.org/10.1109/TNN.2002.806646
Publication status	Published - Jan 2003

Keywords

Neural networks
Optical diffraction
Optimization methods
Packet switching

Access to Document

10.1109/TNN.2002.806646

Cite this

@article{b022dab3fe884de78bb6e33734e3c14c,

title = "A neural-network packet switch controller: Scalability, performance, and network optimization",

abstract = "This paper examines a novel combination of architecture and algorithm for a packet switch controller that incorporates an experimentally implemented optically interconnected neural network. The network performs scheduling decisions based on incoming packet requests and priorities. We show how and why, by means of simulation, the move from a continuous to a discrete algorithm has improved both network performance and scalability. The system's limitations are examined and conclusions drawn as to its maximum scalability and throughput based on today's technologies.",

keywords = "Neural networks, Optical diffraction, Optimization methods, Packet switching",

author = "Symington, {Keith J.} and Waddie, {Andrew J.} and Taghizadeh, {Mohammad R.} and Snowdon, {John F.}",

year = "2003",

month = jan,

doi = "10.1109/TNN.2002.806646",

language = "English",

volume = "14",

pages = "28--34",

journal = "IEEE Transactions on Neural Networks",

publisher = "IEEE",

number = "1",

}

TY - JOUR

T1 - A neural-network packet switch controller

T2 - Scalability, performance, and network optimization

AU - Symington, Keith J.

AU - Waddie, Andrew J.

AU - Taghizadeh, Mohammad R.

AU - Snowdon, John F.

PY - 2003/1

Y1 - 2003/1

N2 - This paper examines a novel combination of architecture and algorithm for a packet switch controller that incorporates an experimentally implemented optically interconnected neural network. The network performs scheduling decisions based on incoming packet requests and priorities. We show how and why, by means of simulation, the move from a continuous to a discrete algorithm has improved both network performance and scalability. The system's limitations are examined and conclusions drawn as to its maximum scalability and throughput based on today's technologies.

AB - This paper examines a novel combination of architecture and algorithm for a packet switch controller that incorporates an experimentally implemented optically interconnected neural network. The network performs scheduling decisions based on incoming packet requests and priorities. We show how and why, by means of simulation, the move from a continuous to a discrete algorithm has improved both network performance and scalability. The system's limitations are examined and conclusions drawn as to its maximum scalability and throughput based on today's technologies.

KW - Neural networks

KW - Optical diffraction

KW - Optimization methods

KW - Packet switching

UR - http://www.scopus.com/inward/record.url?scp=0037276931&partnerID=8YFLogxK

U2 - 10.1109/TNN.2002.806646

DO - 10.1109/TNN.2002.806646

M3 - Article

C2 - 18237987

VL - 14

SP - 28

EP - 34

JO - IEEE Transactions on Neural Networks

JF - IEEE Transactions on Neural Networks

IS - 1

ER -

A neural-network packet switch controller: Scalability, performance, and network optimization

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this