Efficient metacomputing of elliptic linear and non-linear problems

Nicolas Barberou; Marc Garbey; Matthias Hess; Michael M. Resch; Tuomo Rossi; Jari Toivanen; Damien Tromeur-Dervout

doi:10.1016/S0743-7315(03)00003-0

Efficient metacomputing of elliptic linear and non-linear problems

Nicolas Barberou, Marc Garbey, Matthias Hess, Michael M. Resch, Tuomo Rossi, Jari Toivanen, Damien Tromeur-Dervout

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

37 Scopus citations

Abstract

Metacomputing is a method of using the GRID, which originated in the US and quickly was also picked up by European and Japanese researchers where a number of challenging projects were aiming at the exploitation of such distributed resources. Especially the GLOBUS project (The globus project: a status report, Proceedings IPPS/SPDP'98 Heterogeneous Computing Workshop, 1998, pp. 4-18) has contributed to the success of metacomputing substantially. However, high latency and low bandwidth have made people doubt the feasibility of this concept for big simulation codes. In this paper we present new numerical methods that help to exploit such configurations and overcome the problems of low network performance. To proof the feasibility of our approach we show results of simulations in an innovative GRID environment of supercomputers.

Original language	English (US)
Pages (from-to)	564-577
Number of pages	14
Journal	Journal of Parallel and Distributed Computing
Volume	63
Issue number	5
DOIs	https://doi.org/10.1016/S0743-7315(03)00003-0
State	Published - May 1 2003

Keywords

GRID
Metacomputing
Numerical methods
Simulation

ASJC Scopus subject areas

Software
Theoretical Computer Science
Hardware and Architecture
Computer Networks and Communications
Artificial Intelligence

Access to Document

10.1016/S0743-7315(03)00003-0

Cite this

@article{5c7ea1bfa87f463e9329fe3340cd59b1,

title = "Efficient metacomputing of elliptic linear and non-linear problems",

abstract = "Metacomputing is a method of using the GRID, which originated in the US and quickly was also picked up by European and Japanese researchers where a number of challenging projects were aiming at the exploitation of such distributed resources. Especially the GLOBUS project (The globus project: a status report, Proceedings IPPS/SPDP'98 Heterogeneous Computing Workshop, 1998, pp. 4-18) has contributed to the success of metacomputing substantially. However, high latency and low bandwidth have made people doubt the feasibility of this concept for big simulation codes. In this paper we present new numerical methods that help to exploit such configurations and overcome the problems of low network performance. To proof the feasibility of our approach we show results of simulations in an innovative GRID environment of supercomputers.",

keywords = "GRID, Metacomputing, Numerical methods, Simulation",

author = "Nicolas Barberou and Marc Garbey and Matthias Hess and Resch, {Michael M.} and Tuomo Rossi and Jari Toivanen and Damien Tromeur-Dervout",

note = "Funding Information: The authors would like to thank the John von Neuman Institute and the Pittsburgh Supercomputing Center for providing access to their systems for experiments. All authors would like to thank their home organizations Cines, CSC and HLRS for support in this work. The work of Rossi and Toivanen was supported by the Academy of Finland under grants #43066, #53588 and #66407. The work of Nicolas Barberou was supported by the abondement of ANVAR. The work of Tromeur-Dervout was supported by the ACI-GRID of the French Ministry of Research. Copyright: Copyright 2017 Elsevier B.V., All rights reserved.",

year = "2003",

month = may,

day = "1",

doi = "10.1016/S0743-7315(03)00003-0",

language = "English (US)",

volume = "63",

pages = "564--577",

journal = "Journal of Parallel and Distributed Computing",

issn = "0743-7315",

publisher = "Academic Press",

number = "5",

}

TY - JOUR

T1 - Efficient metacomputing of elliptic linear and non-linear problems

AU - Barberou, Nicolas

AU - Garbey, Marc

AU - Hess, Matthias

AU - Resch, Michael M.

AU - Rossi, Tuomo

AU - Toivanen, Jari

AU - Tromeur-Dervout, Damien

N1 - Funding Information: The authors would like to thank the John von Neuman Institute and the Pittsburgh Supercomputing Center for providing access to their systems for experiments. All authors would like to thank their home organizations Cines, CSC and HLRS for support in this work. The work of Rossi and Toivanen was supported by the Academy of Finland under grants #43066, #53588 and #66407. The work of Nicolas Barberou was supported by the abondement of ANVAR. The work of Tromeur-Dervout was supported by the ACI-GRID of the French Ministry of Research. Copyright: Copyright 2017 Elsevier B.V., All rights reserved.

PY - 2003/5/1

Y1 - 2003/5/1

N2 - Metacomputing is a method of using the GRID, which originated in the US and quickly was also picked up by European and Japanese researchers where a number of challenging projects were aiming at the exploitation of such distributed resources. Especially the GLOBUS project (The globus project: a status report, Proceedings IPPS/SPDP'98 Heterogeneous Computing Workshop, 1998, pp. 4-18) has contributed to the success of metacomputing substantially. However, high latency and low bandwidth have made people doubt the feasibility of this concept for big simulation codes. In this paper we present new numerical methods that help to exploit such configurations and overcome the problems of low network performance. To proof the feasibility of our approach we show results of simulations in an innovative GRID environment of supercomputers.

AB - Metacomputing is a method of using the GRID, which originated in the US and quickly was also picked up by European and Japanese researchers where a number of challenging projects were aiming at the exploitation of such distributed resources. Especially the GLOBUS project (The globus project: a status report, Proceedings IPPS/SPDP'98 Heterogeneous Computing Workshop, 1998, pp. 4-18) has contributed to the success of metacomputing substantially. However, high latency and low bandwidth have made people doubt the feasibility of this concept for big simulation codes. In this paper we present new numerical methods that help to exploit such configurations and overcome the problems of low network performance. To proof the feasibility of our approach we show results of simulations in an innovative GRID environment of supercomputers.

KW - GRID

KW - Metacomputing

KW - Numerical methods

KW - Simulation

UR - http://www.scopus.com/inward/record.url?scp=27844573249&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=27844573249&partnerID=8YFLogxK

U2 - 10.1016/S0743-7315(03)00003-0

DO - 10.1016/S0743-7315(03)00003-0

M3 - Article

AN - SCOPUS:27844573249

SN - 0743-7315

VL - 63

SP - 564

EP - 577

JO - Journal of Parallel and Distributed Computing

JF - Journal of Parallel and Distributed Computing

IS - 5

ER -

Efficient metacomputing of elliptic linear and non-linear problems

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this