High performance computing on low cost computers: a review of parallel and distributed computing methodologies for finite element analysis

Ian Mackie

High performance computing on low cost computers: a review of parallel and distributed computing methodologies for finite element analysis

Ian Mackie

Research output: Chapter in Book/Report/Conference proceeding › Chapter

Abstract

The paper considers three main software technologies for implementing parallelism: MPI, .NET and Java. Attention is given to the role of object and component oriented program design methods. Consideration is given to the mathematical methods used, with special emphasis on the use of domain decomposition methods. The interplay between software engineering, mathematical methods used, and user interaction is examined. There is a particular focus on the role of object and component oriented programming methods, and how they facilitate the use of domain decomposition methods and integrating these with machine architectures and implementation of user-interaction features. Some work on the forthcoming Task Parallel Library (TPL) for .NET is presented.
Speed comparisons are presented for C++ and MPI, .NET and Java. Overall C++ with MPI is faster than .NET or Java, but the difference is not that great to rule out the use of .NET or Java.

Original language	English
Title of host publication	Developments and applications in engineering computational technology
Editors	B.H.V. Topping, J.M. Adam, F.J. Pallares, R. Bru, M.L. Romero
Place of Publication	Kippen
Publisher	Saxe-Coburg Publications
Pages	263-284
Number of pages	22
ISBN (Print)	9781874672487
Publication status	Published - 2010
Event	Seventh International Conference on Engineering Computational Technology - Universidad Politécnica de Valencia, Valencia, Spain Duration: 14 Sept 2010 → 17 Sept 2010 http://www.civil-comp.com/conf/ect2010.htm

Publication series

Name	Computational Science, Engineering and Technology
Publisher	Saxe-Coburg
Volume	26
ISSN (Print)	1759-3158

Conference

Conference	Seventh International Conference on Engineering Computational Technology
Abbreviated title	ECT 2010
Country/Territory	Spain
City	Valencia
Period	14/09/10 → 17/09/10
Internet address	http://www.civil-comp.com/conf/ect2010.htm

Keywords

Distributed computing
Finite element analysis
Object-oriented programming
Parallel processing
Component oriented

Access to Document

http://www.saxe-coburg.co.uk/pubs/pub.html?id=26

Cite this

Mackie, I. (2010). High performance computing on low cost computers: a review of parallel and distributed computing methodologies for finite element analysis. In B. H. V. Topping, J. M. Adam, F. J. Pallares, R. Bru, & M. L. Romero (Eds.), Developments and applications in engineering computational technology (pp. 263-284). (Computational Science, Engineering and Technology; Vol. 26). Saxe-Coburg Publications. http://www.saxe-coburg.co.uk/pubs/pub.html?id=26

Mackie, Ian. / High performance computing on low cost computers : a review of parallel and distributed computing methodologies for finite element analysis. Developments and applications in engineering computational technology. editor / B.H.V. Topping ; J.M. Adam ; F.J. Pallares ; R. Bru ; M.L. Romero. Kippen : Saxe-Coburg Publications, 2010. pp. 263-284 (Computational Science, Engineering and Technology).

@inbook{6eb4c8d648944c9da7b062edc68079d3,

title = "High performance computing on low cost computers: a review of parallel and distributed computing methodologies for finite element analysis",

abstract = "The paper considers three main software technologies for implementing parallelism: MPI, .NET and Java. Attention is given to the role of object and component oriented program design methods. Consideration is given to the mathematical methods used, with special emphasis on the use of domain decomposition methods. The interplay between software engineering, mathematical methods used, and user interaction is examined. There is a particular focus on the role of object and component oriented programming methods, and how they facilitate the use of domain decomposition methods and integrating these with machine architectures and implementation of user-interaction features. Some work on the forthcoming Task Parallel Library (TPL) for .NET is presented.Speed comparisons are presented for C++ and MPI, .NET and Java. Overall C++ with MPI is faster than .NET or Java, but the difference is not that great to rule out the use of .NET or Java.",

keywords = "Distributed computing, Finite element analysis, Object-oriented programming, Parallel processing, Component oriented",

author = "Ian Mackie",

year = "2010",

language = "English",

isbn = "9781874672487",

series = "Computational Science, Engineering and Technology",

publisher = "Saxe-Coburg Publications",

pages = "263--284",

editor = "B.H.V. Topping and J.M. Adam and F.J. Pallares and R. Bru and M.L. Romero",

booktitle = "Developments and applications in engineering computational technology",

note = "Seventh International Conference on Engineering Computational Technology , ECT 2010 ; Conference date: 14-09-2010 Through 17-09-2010",

url = "http://www.civil-comp.com/conf/ect2010.htm",

}

Mackie, I 2010, High performance computing on low cost computers: a review of parallel and distributed computing methodologies for finite element analysis. in BHV Topping, JM Adam, FJ Pallares, R Bru & ML Romero (eds), Developments and applications in engineering computational technology. Computational Science, Engineering and Technology, vol. 26, Saxe-Coburg Publications, Kippen, pp. 263-284, Seventh International Conference on Engineering Computational Technology , Valencia, Spain, 14/09/10. <http://www.saxe-coburg.co.uk/pubs/pub.html?id=26>

High performance computing on low cost computers: a review of parallel and distributed computing methodologies for finite element analysis. / Mackie, Ian.
Developments and applications in engineering computational technology. ed. / B.H.V. Topping; J.M. Adam; F.J. Pallares; R. Bru; M.L. Romero. Kippen: Saxe-Coburg Publications, 2010. p. 263-284 (Computational Science, Engineering and Technology; Vol. 26).

Research output: Chapter in Book/Report/Conference proceeding › Chapter

TY - CHAP

T1 - High performance computing on low cost computers

T2 - Seventh International Conference on Engineering Computational Technology

AU - Mackie, Ian

PY - 2010

Y1 - 2010

N2 - The paper considers three main software technologies for implementing parallelism: MPI, .NET and Java. Attention is given to the role of object and component oriented program design methods. Consideration is given to the mathematical methods used, with special emphasis on the use of domain decomposition methods. The interplay between software engineering, mathematical methods used, and user interaction is examined. There is a particular focus on the role of object and component oriented programming methods, and how they facilitate the use of domain decomposition methods and integrating these with machine architectures and implementation of user-interaction features. Some work on the forthcoming Task Parallel Library (TPL) for .NET is presented.Speed comparisons are presented for C++ and MPI, .NET and Java. Overall C++ with MPI is faster than .NET or Java, but the difference is not that great to rule out the use of .NET or Java.

AB - The paper considers three main software technologies for implementing parallelism: MPI, .NET and Java. Attention is given to the role of object and component oriented program design methods. Consideration is given to the mathematical methods used, with special emphasis on the use of domain decomposition methods. The interplay between software engineering, mathematical methods used, and user interaction is examined. There is a particular focus on the role of object and component oriented programming methods, and how they facilitate the use of domain decomposition methods and integrating these with machine architectures and implementation of user-interaction features. Some work on the forthcoming Task Parallel Library (TPL) for .NET is presented.Speed comparisons are presented for C++ and MPI, .NET and Java. Overall C++ with MPI is faster than .NET or Java, but the difference is not that great to rule out the use of .NET or Java.

KW - Distributed computing

KW - Finite element analysis

KW - Object-oriented programming

KW - Parallel processing

KW - Component oriented

M3 - Chapter

SN - 9781874672487

T3 - Computational Science, Engineering and Technology

SP - 263

EP - 284

BT - Developments and applications in engineering computational technology

A2 - Topping, B.H.V.

A2 - Adam, J.M.

A2 - Pallares, F.J.

A2 - Bru, R.

A2 - Romero, M.L.

PB - Saxe-Coburg Publications

CY - Kippen

Y2 - 14 September 2010 through 17 September 2010

ER -

Mackie I. High performance computing on low cost computers: a review of parallel and distributed computing methodologies for finite element analysis. In Topping BHV, Adam JM, Pallares FJ, Bru R, Romero ML, editors, Developments and applications in engineering computational technology. Kippen: Saxe-Coburg Publications. 2010. p. 263-284. (Computational Science, Engineering and Technology).

High performance computing on low cost computers: a review of parallel and distributed computing methodologies for finite element analysis

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this