nD-RAPID: A multidimensional scalable fault-tolerant optoelectronic interconnection for high-performance computing systems

Chander Kochar, Avinash Kodi, Ahmed Louri

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

The increasing demand for bandwidth coupled with saturating electrical systems is leading the drive for optics as an interconnect technology. High-performance computing systems (HPCS) consist of a large number of components such as processors, memories, and interconnection links. As the number of components in HPCS increases, the probability of a failure also increases. Therefore, it becomes imperative that the system be fault tolerant to ensure high availability even in the presence of faults. We propose a multidimensional optoelectronic architecture, nD-RAPID (reconfigurable and scalable allphotonic interconnect for distributed and parallel systems), where n can be 1, 2, or 3. nD-RAPID provides high bandwidth, low latency, dynamic reconfiguration, and fault tolerance. While designing the fault-tolerant routing algorithm, we have tried to ensure that it provides optimum performance in the absence of faults, shows minimal degradation in the presence of faults, and can tolerate a reasonable number of faults. In the presence of faults the onboard switching mechanism dynamically reconfigures itself to reroute packets along nonfaulty links. Extensive simulation results are presented that compare nD-RAPID with other popular HPCS topologiesr.

Original languageEnglish (US)
Pages (from-to)465-481
Number of pages17
JournalJournal of Optical Networking
Volume6
Issue number5
DOIs
StatePublished - May 1 2007

Fingerprint

Optoelectronic devices
Bandwidth
Routing algorithms
Fault tolerance
Optics
Availability
bandwidth
fault tolerance
Data storage equipment
Degradation
availability
central processing units
optics
degradation
simulation

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Atomic and Molecular Physics, and Optics

Cite this

nD-RAPID : A multidimensional scalable fault-tolerant optoelectronic interconnection for high-performance computing systems. / Kochar, Chander; Kodi, Avinash; Louri, Ahmed.

In: Journal of Optical Networking, Vol. 6, No. 5, 01.05.2007, p. 465-481.

Research output: Contribution to journalArticle

@article{6beae525c77b41c4bfb5bb3190a151bd,
title = "nD-RAPID: A multidimensional scalable fault-tolerant optoelectronic interconnection for high-performance computing systems",
abstract = "The increasing demand for bandwidth coupled with saturating electrical systems is leading the drive for optics as an interconnect technology. High-performance computing systems (HPCS) consist of a large number of components such as processors, memories, and interconnection links. As the number of components in HPCS increases, the probability of a failure also increases. Therefore, it becomes imperative that the system be fault tolerant to ensure high availability even in the presence of faults. We propose a multidimensional optoelectronic architecture, nD-RAPID (reconfigurable and scalable allphotonic interconnect for distributed and parallel systems), where n can be 1, 2, or 3. nD-RAPID provides high bandwidth, low latency, dynamic reconfiguration, and fault tolerance. While designing the fault-tolerant routing algorithm, we have tried to ensure that it provides optimum performance in the absence of faults, shows minimal degradation in the presence of faults, and can tolerate a reasonable number of faults. In the presence of faults the onboard switching mechanism dynamically reconfigures itself to reroute packets along nonfaulty links. Extensive simulation results are presented that compare nD-RAPID with other popular HPCS topologiesr.",
author = "Chander Kochar and Avinash Kodi and Ahmed Louri",
year = "2007",
month = "5",
day = "1",
doi = "10.1364/JON.6.000465",
language = "English (US)",
volume = "6",
pages = "465--481",
journal = "Journal of Optical Networking",
issn = "1536-5379",
publisher = "The Optical Society",
number = "5",

}

TY - JOUR

T1 - nD-RAPID

T2 - A multidimensional scalable fault-tolerant optoelectronic interconnection for high-performance computing systems

AU - Kochar, Chander

AU - Kodi, Avinash

AU - Louri, Ahmed

PY - 2007/5/1

Y1 - 2007/5/1

N2 - The increasing demand for bandwidth coupled with saturating electrical systems is leading the drive for optics as an interconnect technology. High-performance computing systems (HPCS) consist of a large number of components such as processors, memories, and interconnection links. As the number of components in HPCS increases, the probability of a failure also increases. Therefore, it becomes imperative that the system be fault tolerant to ensure high availability even in the presence of faults. We propose a multidimensional optoelectronic architecture, nD-RAPID (reconfigurable and scalable allphotonic interconnect for distributed and parallel systems), where n can be 1, 2, or 3. nD-RAPID provides high bandwidth, low latency, dynamic reconfiguration, and fault tolerance. While designing the fault-tolerant routing algorithm, we have tried to ensure that it provides optimum performance in the absence of faults, shows minimal degradation in the presence of faults, and can tolerate a reasonable number of faults. In the presence of faults the onboard switching mechanism dynamically reconfigures itself to reroute packets along nonfaulty links. Extensive simulation results are presented that compare nD-RAPID with other popular HPCS topologiesr.

AB - The increasing demand for bandwidth coupled with saturating electrical systems is leading the drive for optics as an interconnect technology. High-performance computing systems (HPCS) consist of a large number of components such as processors, memories, and interconnection links. As the number of components in HPCS increases, the probability of a failure also increases. Therefore, it becomes imperative that the system be fault tolerant to ensure high availability even in the presence of faults. We propose a multidimensional optoelectronic architecture, nD-RAPID (reconfigurable and scalable allphotonic interconnect for distributed and parallel systems), where n can be 1, 2, or 3. nD-RAPID provides high bandwidth, low latency, dynamic reconfiguration, and fault tolerance. While designing the fault-tolerant routing algorithm, we have tried to ensure that it provides optimum performance in the absence of faults, shows minimal degradation in the presence of faults, and can tolerate a reasonable number of faults. In the presence of faults the onboard switching mechanism dynamically reconfigures itself to reroute packets along nonfaulty links. Extensive simulation results are presented that compare nD-RAPID with other popular HPCS topologiesr.

UR - http://www.scopus.com/inward/record.url?scp=40249091437&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=40249091437&partnerID=8YFLogxK

U2 - 10.1364/JON.6.000465

DO - 10.1364/JON.6.000465

M3 - Article

AN - SCOPUS:40249091437

VL - 6

SP - 465

EP - 481

JO - Journal of Optical Networking

JF - Journal of Optical Networking

SN - 1536-5379

IS - 5

ER -