Catálogo de publicaciones - libros

We study the scaling properties of the quantum projected (5) model in three dimensions by means of a highly accurate Quantum-Monte-Carlo analysis. Within the parameter regime studied (temperature and system size), we show that the scaling behavior is consistent with a (5)-symmetric critical behavior in the numerically accessible region. This holds both when the symmetry breaking is caused by quantum fluctuations only as well as when also the static (mean-field) symmetry is moderately broken. We argue that possible departure away from the (5) - symmetric scaling occurs only in an extremely narrow parameter regime, which is inaccessible both experimentally and numerically.

More reliable turbulence models are urgently needed for the aircraft, automobile and chemical industries. The development of such models needs new concepts and the validation by complete and accurate data sets. In this work a novel Reynolds stress model has been developed based on invariant theory and two-point correlation technique. This model is extensively tested both via and testing based on available data sets. For the dissipation rate equation, the existing data suffered from lack of spatial resolution. So highly resolved direct numerical simulations have been carried out to shed more light onto the processes described by that equation. In addition, the novel model was implemented in engineering software, to make the new developments available for technical applications.

We have developed a simulation environment for the efficient numerical computation of flow induced sound. Thereby, the fluid flow program FASTEST-3D has been coupled via MpCCI to CFS++ (Coupled Field Simulation), which performs the sound field computation. Thereby, different computational domains as well as grids for the fluid field and acoustic field can be chosen. As an practical example, we discuss the computation of the emitted noise from a square cylinder within a turbulent flow.

In this contribution, fully three-dimensional models are used for the numerical simulation of both the structure and the fluid in fluid-structure interaction computations. A partitioned, but fully implicit coupling algorithm is employed. As an example, the wind-excitation of a thin-walled tower is investigated.

The prediction of MOVPE processes requires the modeling of numerous coupled transport phenomena for momentum, mass and heat including temperature dependent physical properties and chemical reactions. In the present paper, the numerical simulation is used as a tool to identify and distinguish systematically the chemical parameters in the AlGaN growth process in order to obtain a reliable and efficient prediction of the process.

This work presents a parallel Monte Carlo algorithm for the calculation of combined radiative and conductive heat transfer. The proposed formulation effectively separates the time-consuming ray-tracing part of the Monte Carlo method from the energy computations required in the iterative solution of the energy equation. The method is applied for a simple combined radiative and conductive heat transfer problem and excellent agreement with the benchmark results is found. The ray-tracing part of the algorithm is parallelised and applied in two configurations, which represent the opposite ends of the currently available parallel computer architectures; a PC cluster and the Hitachi SR8000-F1 supercomputer. For sufficiently large sampling sets, the measurements show an almost ideal speedup.

Facing the problem of implementing an efficient solver for partial differential equations, we are, in general, confronted with a certain quandary between numerical efficiency and efficiency in the usage of hardware resources: Modern numerical methods require the handling of hierarchical multilevel data on adaptively refined data structures, which are mostly represented by trees. On the other hand, as data access is one of the most important bottlenecks in high performance computing, we would wish to process data linearly with a high locality in time and space to be able to exploit the capability of cache hierarchies. In this paper, we show an approach based on space-filling curves as an odering mechanism for the cells of space-tree grids, with the help of which we can transform our (inherently highly non-local) data respresentation by trees to a few linearly processed data sets. As a consequence, we reach extremely high cache hit-rates above 99, 9%. In addition, the used methods make both parallelization and multigrid algorithms on adaptive grids with hierarchical data very straightforward and efficient.

This article presents a computational steering research project coupling a supercomputer with a virtual reality (VR) environment to allow for interaction during a real-time CFD simulation using an immersive high-end visualization interface. Interaction comprises not only the changing of parameters, but also the modification of geometry, e.g. removing, adding and transforming objects in a virtual CAD-generated room during runtime.

The underlying CFD computation and grid generation is processed on a high-performance computer (HPC) to enable a real-time simulation which instantly reacts to user manipulations. Based on the Lattice-Boltzmann method, the simulation kernel shows good parallel efficiency on the Hitachi SR8000 pseudo-vector supercomputer at the Leibniz Computing Center (LRZ) in Munich. For post-processing and steering in virtual reality a stereoscopic projection screen and a tracked wand input device is used. To achieve optimal immerson a head-tracked view is supported and complemented by a context-sensitive 3D VR menu.

ByGRID is a Grid project based on the GLOBUS toolkit. Its purpose is to (i) demonstrate the functionality of Grid middleware, delivering to end users the benefits, but not the complexity of Grid-based workflow and to (ii) share, for the benefit of the users, unique resources between the two computing centers involved, RRZE and LRZ. We examine ByGRID from the computing center’s view, reporting on setup, encountered problems, and administration, and from the user’s point of view, explaining a practical usage scenario

This report describes the exploration of the possibilities for the direct synthesis of concurrent C programs on high-performance computers in FPGAs