Adaptive Resolution Data Access For Visualization of Magnetohydrodynamic Simulation Data

Interactive visualization of very large data sets remains a challenging problem to the visualization community. One promising solution involves using adaptive resolution representations of the data. In this model, important regions of data can be selected using either error analysis or by allowing the scientist to interactively zoom to important phenomena. Regions of interest are rendered in high resolution data, while low and/or medium resolution representations are used to render regions in which the data does not vary much. Our data comes from numerical simulation of the solar wind with the Earth’s magnetosphere. We have developed software tools and support libraries to aid in the interactive visualization of large data sets by using our adaptive resolution data model. We show that with this software, we can achieve significant performance gains while maintaining good image quality and maintain interactive frame rates.