Jacek A. Jankowski :: Site|
Jacek A. Jankowski :: Site |
||
| position ::: publications / UnTRIM / | ||
U n T R I M - connected publications
Abstract. This document is aimed to provide the user of the MPI version of UnTRIM with necessary information for the installation, compilation and especially execution of the software.
The text can be obtained from the BAW site.
|
Reference: Mathematical Model UnTRIM - MPI Version Manual, BAW, Technical Report, Karlsruhe 2008. |
 |
J.A. Jankowski
Abstract.This contribution deals with further developments of UnTRIM, an unstructured-grid, three-dimensional, semi-implicit finite difference finite volume model for the shallow water equations (Casulli, 2002). Attractive numerical properties of the method like its robustness due to the unlimited stability and successes in practical applications spawned efforts aimed at making the available code fit for the high performance computing in order to address larger, complex problems in hydraulic engineering. The paper concentrates on the parallel implementation of the program, based on the domain decomposition method and message passing, which has been achieved without negatively affecting any of the properties of the serial code. A special attention is paid to a new, autonomous parallel streamline backtracking algorithm, which allows using semi-Lagrangian methods in decomposed meshes without compromising the scalability of the code. The new developments have been carefully verified not only with the numerous simple, abstract test cases illustrating the application domain of the code, but also with advanced, high resolution models presently applied for research and engineering projects.
The text can be obtained locally.
|
Reference: Jankowski, J.A. (2007) Further developments of UnTRIM:
parallel implementation and its verification.
Proceedings of the |
|
J.A. Jankowski
Abstract. The parallel implementation of an unstructured-grid, three-dimensional, semi-implicit finite difference and finite volume model for the free surface Navier-Stokes equations (UnTRIM) is presented and discussed. The new developments are aimed to make the code available for high performance computing in order to address larger, complex problems in environmental free surface flows. The parallelisation is based on the mesh partitioning method and message passing and has been achieved without negatively affecting any of the advantageous properties of the serial code, like its robustness, accuracy and efficiency. The key issue is a new, autonomous parallel streamline backtracking algorithm, which allows using semi-Lagrangian methods in decomposed meshes without compromising the scalability of the code. The implementation has been carefully verified not only with simple, abstract test cases illustrating the application domain of the code, but also with advanced, high resolution models presently applied for research and engineering projects. The scheme performance and accuracy aspects are researched and discussed.
|
Reference: Jankowski, J.A. (2009) Parallel implementation of a
non-hydrostatic model for free surface flows with semi-Lagrangian advection
treatment. |
|
R. Patzwahl, J.A. Jankowski, T. Lege
Abstract. The paper discusses the advantages and disadvantages of numerical modelling of rivers with a grid of very high resolution using the new MPI version of UnTRIM as the computational engine. The main aim of the investigation is to assess the economy of this approach taking especially into account the effort required for the mesh generation and its modification, which is usually laborious in the case of coarser meshes requiring exact reproduction of structure lines defining the flow. In the first step the results of a low-resolution Telemac-2D grid are compared to the results obtained for the same river stretch topography with a high-resolution UnTRIM model. In the second step the results of two- and three-dimensional modelling applying a high-resolution mesh based on a high quality digital terrain model are studied. It is concluded that under the assumption of appropriate computational resources readily available, the high-resolution modelling reduces significantly the effort required for the initial model set-up and for adjustments due to changes in the model topography. The calibration, parametrisation and validation of the models is simplified without affecting the accuracy, freeing the engineer to concentrate on the project aims and not bypass the weak points of the methodology.
The text can be obtained locally.
|
Reference: R. Patzwahl, J.A. Jankowski, T. Lege (2008) Very high resolution numerical modelling for inland waterway design. In: International Conference on Fluvial Hydraulics (River Flow 2008), Izmir, Turkey, 2008. |
|
J.A. Jankowski
Abstract.The rapid changes in the present computer architectures throw questions concerning the future of the hardware-oriented programming based on the usage of heterogeneous parallel computing resources, consisting of nodes of interconnected, but separate CPUs and GPUs. However, it is believed that the sustainability of the programming efforts nowadays is to be searched in the fact that the principle source of the performance of the present GPUs as well as the emerging APUs is the hardware acceleration of vector operations. Therefore, this contribution delivers an assessment of a hardware-accelerated parallel implementation of a principal algorithm for a vertically integrated finite difference scheme for free surface flows including non-linear treatment of wetting and drying. The chosen approach is to expose the fine-grained parallelism of the scheme and execute the whole computational kernel of the code on a presently available streaming vector processor – a state-of-the-art GPU. The reached speedups compared to a single CPU core are in the order 20 or 30 for the double or single precision, respectively, which opens new possibilities for high resolution flood modelling. However, in order to reach this speedup level the coding must be radically adapted for the vectorised execution.
The preprint text can be obtained locally.
|
Reference: J.A. Jankowski(2011) Assessment of a parallel implementation of a two-dimensional scheme for free surface flows using a GPU Unpublished. |
|
 |
|
| © 2001 by jaj • mail: jacek.jankowski@web.de | |