Página 1 dos resultados de 1 itens digitais encontrados em 0.002 segundos

High-Performance Computing in Geoscience - Data Preprocessing by Domain Decomposition and Load Balancing; High-Performance Computing in den Geowissenschaften - Datenvorverarbeitung mittels Gebietszerlegung und Lastbalancierung

Kemmler, Dany
Fonte: Universität Tübingen Publicador: Universität Tübingen
Tipo: Dissertation; info:eu-repo/semantics/doctoralThesis
Relevância na Pesquisa
The popularity and availability of computers is a simple fact of life in most of today's world. Computer simulation is a quick and relatively inexpensive alternative to physical experimentation in the scientific realm. Computer operations are meanwhile performed increasingly rapidly; one trillion operations per second are not anything extraordinary these days and are necessary in the field of automotive engineering, weather forecast or applied geology, for example. An interdependency of speed, power and storage space means that if one of these attributes is insufficient or impaired, it will limit the efficacy of the others. Data preprocessing and problem subdivision play a vital role in making real-world problems "computable." High-performance computing is not only a cornerstone of modern life, but also a compelling topic in itself. The task of descretizing real-world problems and processing models so that these problems can be solved by computers involves obtaining finite amounts of data from real-world problem domains and replacing them by grids consisting of inter-connected nodes and elements which will serve to model the problem in question on a computer. Altering the size of the grid, its number of nodes, etc. to determine the optimal structure to simulate a certain problem on a computer is the ultimate goal here. The more carefully discretization is accomplished...