Visualisation of large unstructured grids within a parallel framework
Jason W. Jones, Nigel P. Weatherill · 14th Computational Fluid Dynamics Conference · 1999
4. Computational Analysis 5. Solution Interrogation and A graphical environment is presented in which the entire computational simulation process may be performed using parallel processing techniques. As computational analysis techniques have matured, their use has spread rapidly beyond the research environment into industry in which they are used purely as a tool in order to obtain the required results. Allowing the traditional engineer easy access to these tools requires that the complex numerical algorithms be hidden behind a user-friendly, graphically interactive environment. This expansion into industry has also led to a requirement to simulate real-world problems involving large and complex geometries which, in turn, demand a large increase in the size and complexity of the problem which can be tackled. To meet this requirement,-the need to distribute both the data and the computational effort amongst multiple processors (or computers) becomes evident. The aim’ of the presented graphical environment is to allow the interaction with, and solution of problemsinvolving complex geometries (involving several thousand surface patches) and large unstructured meshes (of the order of 20-100 million elements) with no sequential bottlenecks at any stage of the process. 6. Mesh Refinement. The Specification phase is where the user interacts with a geometry and specifies all the required input data, such as grid point density parameters. Recently, the specification phase has utilised advanced graphics workstations, and algorithms embedded within easy+o-use Graphical User Interfaces (GUIs) . The Grid Generation phase is where, given all the required input parameters, the grid is generated using relatively high compute intensive algorithms. Usually, there is little user interaction at this stage. The Grid Evaluation phase involves the user evaluating the suitability and quality of thegenerated grid. This often requires a high degree of user interaction using statistical analysis of grid quality and visualisation of the regions of-the domain which contain poorly formed elements. Again, in the Grid Evaluation phase the role of the GUI is important.