On the scalability of parallel computations on a network of workstations

Domain decomposition (DD) is confirmed to be a useful parallel technique for parallel simulation. Mesh partition strategy together with a master/slave construction represents a natural way of doing parallel distributed computations. In a recent study using an unstructured finite volume implementatio...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Computational mechanics Jg. 22; H. 4; S. 344 - 354
Hauptverfasser: Dou, H.-S., Phan-Thien, N.
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Heidelberg Springer 01.10.1998
Berlin Springer Nature B.V
Schlagworte:
Algorithmics. Computability. Computer arithmetics
Applied sciences
Computational methods in fluid dynamics
Computer science; control theory; systems
Computer simulation
Exact sciences and technology
Finite element method
Fluid dynamics
Fundamental areas of phenomenology (including applications)
Physics
Theoretical computing
Virtual environments
Workstations
Viscoelastic fluid
Computational fluid dynamics
Computer network
Numerical simulation
Parallel computation
Incompressible fluid
Flow field
Domain decomposition
Numerical convergence
Computability
Flat plate
ISSN:0178-7675, 1432-0924
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Domain decomposition (DD) is confirmed to be a useful parallel technique for parallel simulation. Mesh partition strategy together with a master/slave construction represents a natural way of doing parallel distributed computations. In a recent study using an unstructured finite volume implementation with Parallel Virtual Machine (PVM), Dou and Phan-Thien [1] found that there is a limit to the speed-up when the problem size increases to a certain value. In this paper, we will discuss the parallel strategies for DD technique, the load balancing, and the communication issue among the processors, and explain the mechanism of the limit, and search for the way to enhance the speed up and the scalability.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0178-7675
1432-0924
DOI:10.1007/s004660050366