HOBONET

AN INTER-PU CONNECTION NETWORK WITH FAULT-TOLERANCY.

Gyo Osawa, Takashi Yokota, Hideharu Amano, Hideo Aiso

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The authors propose a parallel machine, called HOBONET, in which the PUs are connected with a nearest neighbor mesh (NNM) connection structure. HOBONET consists of a set of switching modules (SwMs) for the purpose of interconnecting the PUs. With the detection of faulty PUs, connections of SwMs are changed so that faulty PUs are disconnected, but the NNM connection structure is maintained logically. The scheme is called lattice defect recovery.

Original languageEnglish
Title of host publicationProceedings of the International Conference on Parallel Processing
EditorsRobert M. Keller
PublisherIEEE
Pages165-168
Number of pages4
ISBN (Print)081860560X
Publication statusPublished - 1984

Fingerprint

Crystal defects
Recovery

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Osawa, G., Yokota, T., Amano, H., & Aiso, H. (1984). HOBONET: AN INTER-PU CONNECTION NETWORK WITH FAULT-TOLERANCY. In R. M. Keller (Ed.), Proceedings of the International Conference on Parallel Processing (pp. 165-168). IEEE.

HOBONET : AN INTER-PU CONNECTION NETWORK WITH FAULT-TOLERANCY. / Osawa, Gyo; Yokota, Takashi; Amano, Hideharu; Aiso, Hideo.

Proceedings of the International Conference on Parallel Processing. ed. / Robert M. Keller. IEEE, 1984. p. 165-168.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Osawa, G, Yokota, T, Amano, H & Aiso, H 1984, HOBONET: AN INTER-PU CONNECTION NETWORK WITH FAULT-TOLERANCY. in RM Keller (ed.), Proceedings of the International Conference on Parallel Processing. IEEE, pp. 165-168.
Osawa G, Yokota T, Amano H, Aiso H. HOBONET: AN INTER-PU CONNECTION NETWORK WITH FAULT-TOLERANCY. In Keller RM, editor, Proceedings of the International Conference on Parallel Processing. IEEE. 1984. p. 165-168
Osawa, Gyo ; Yokota, Takashi ; Amano, Hideharu ; Aiso, Hideo. / HOBONET : AN INTER-PU CONNECTION NETWORK WITH FAULT-TOLERANCY. Proceedings of the International Conference on Parallel Processing. editor / Robert M. Keller. IEEE, 1984. pp. 165-168
@inproceedings{0fae9ddf4a7840bfbc8984b032e246a8,
title = "HOBONET: AN INTER-PU CONNECTION NETWORK WITH FAULT-TOLERANCY.",
abstract = "The authors propose a parallel machine, called HOBONET, in which the PUs are connected with a nearest neighbor mesh (NNM) connection structure. HOBONET consists of a set of switching modules (SwMs) for the purpose of interconnecting the PUs. With the detection of faulty PUs, connections of SwMs are changed so that faulty PUs are disconnected, but the NNM connection structure is maintained logically. The scheme is called lattice defect recovery.",
author = "Gyo Osawa and Takashi Yokota and Hideharu Amano and Hideo Aiso",
year = "1984",
language = "English",
isbn = "081860560X",
pages = "165--168",
editor = "Keller, {Robert M.}",
booktitle = "Proceedings of the International Conference on Parallel Processing",
publisher = "IEEE",

}

TY - GEN

T1 - HOBONET

T2 - AN INTER-PU CONNECTION NETWORK WITH FAULT-TOLERANCY.

AU - Osawa, Gyo

AU - Yokota, Takashi

AU - Amano, Hideharu

AU - Aiso, Hideo

PY - 1984

Y1 - 1984

N2 - The authors propose a parallel machine, called HOBONET, in which the PUs are connected with a nearest neighbor mesh (NNM) connection structure. HOBONET consists of a set of switching modules (SwMs) for the purpose of interconnecting the PUs. With the detection of faulty PUs, connections of SwMs are changed so that faulty PUs are disconnected, but the NNM connection structure is maintained logically. The scheme is called lattice defect recovery.

AB - The authors propose a parallel machine, called HOBONET, in which the PUs are connected with a nearest neighbor mesh (NNM) connection structure. HOBONET consists of a set of switching modules (SwMs) for the purpose of interconnecting the PUs. With the detection of faulty PUs, connections of SwMs are changed so that faulty PUs are disconnected, but the NNM connection structure is maintained logically. The scheme is called lattice defect recovery.

UR - http://www.scopus.com/inward/record.url?scp=0021546115&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0021546115&partnerID=8YFLogxK

M3 - Conference contribution

SN - 081860560X

SP - 165

EP - 168

BT - Proceedings of the International Conference on Parallel Processing

A2 - Keller, Robert M.

PB - IEEE

ER -