Gyrokinetic Particle Simulation of Turbulent Transport in Burning Plasmas [electronic resource].
- Washington, D.C. : United States. Dept. of Energy, 2011. and Oak Ridge, Tenn. : Distributed by the Office of Scientific and Technical Information, U.S. Dept. of Energy.
- Physical Description:
- 4 pages : digital, PDF file
- Additional Creators:
- University of California, Davis, United States. Department of Energy, and United States. Department of Energy. Office of Scientific and Technical Information
- In this project, we have developed techniques for visualizing large-scale time-varying multivariate particle and field data produced by the GPS_TTBP team. Our basic approach to particle data visualization is to provide the user with an intuitive interactive interface for exploring the data. We have designed a multivariate filtering interface for scientists to effortlessly isolate those particles of interest for revealing structures in densely packed particles as well as the temporal behaviors of selected particles. With such a visualization system, scientists on the GPS-TTBP project can validate known relationships and temporal trends, and possibly gain new insights in their simulations. We have tested the system using over several millions of particles on a single PC. We will also need to address the scalability of the system to handle billions of particles using a cluster of PCs. To visualize the field data, we choose to use direct volume rendering. Because the data provided by PPPL is on a curvilinear mesh, several processing steps have to be taken. The mesh is curvilinear in nature, following the shape of a deformed torus. Additionally, in order to properly interpolate between the given slices we cannot use simple linear interpolation in Cartesian space but instead have to interpolate along the magnetic field lines given to us by the scientists. With these limitations, building a system that can provide an accurate visualization of the dataset is quite a challenge to overcome. In the end we use a combination of deformation methods such as deformation textures in order to fit a normal torus into their deformed torus, allowing us to store the data in toroidal coordinates in order to take advantage of modern GPUs to perform the interpolation along the field lines for us. The resulting new rendering capability produces visualizations at a quality and detail level previously not available to the scientists at the PPPL. In summary, in this project we have successfully created new capabilities for the scientists to visualize their 3D data at higher accuracy and quality, enhancing their ability to evaluate the simulations and understand the modeled phenomena.
- Published through SciTech Connect., 12/21/2011., "doe-davis-54956-1", and Ma, Kwan-Liu.
- Type of Report and Period Covered Note:
- Final; 03/01/2008 - 02/28/2011
- Funding Information:
View MARC record | catkey: 14736933