Computational & Technology Resources
an online resource for computational,
engineering & technology publications |
|
Computational Science, Engineering & Technology Series
ISSN 1759-3158 CSETS: 9
COMPUTATIONAL MECHANICS USING HIGH PERFORMANCE COMPUTING Edited by: B.H.V. Topping
Chapter 1
Distributed Visualization for Scientific Simulations H. Katz+ and L. Lämmer*
+Institut für Numerische Methoden und Informatik im Bauwesen, Technische Universität Darmstadt, Darmstadt, Germany H. Katz, L. Lämmer, "Distributed Visualization for Scientific Simulations", in B.H.V. Topping, (Editor), "Computational Mechanics using High Performance Computing", Saxe-Coburg Publications, Stirlingshire, UK, Chapter 1, pp 1-28, 2002. doi:10.4203/csets.9.1
Abstract
The application of the finite element method is popular in both
science and engineering. The use of parallel computers has
significantly increased the size and complexity of problems that can be
solved. Finite element calculations in parallel environments are
subject to very broad research activity and therefore a significant
number of software solutions
are available for a large variety of problem classes. This development is driven by
the increasing availability of parallel computers, especially of networked
workstation and PC clusters.
The result of a finite element calculation is a large amount of numerical data that has to be judged and evaluated. The most efficient way is to visualize the data. In most cases, this is done as a postprocess to the calculation itself. Some existing solutions to this task will exemplarily be presented. Subsequently, the development of a visualization application will be discussed. Our solution has been developed to overcome the bottleneck of the postprocess visualization. In order to allow visualization of data simultaneously with the parallel calculation process several aspects have to be taken into account: an efficient data structure; fast mechanisms for communication between the calculation processes and the visualization application; algorithms to handle large data sets; and methods for data reduction. purchase the full-text of this chapter (price £20)
go to the previous chapter |
|