Parallel ray tracing in scientific visualization

Ray tracing presents an efficient rendering algorithm for scientific visualization using common visualization tools and scales with increasingly large geometry counts while allowing for accurate physically-based visualization and analysis, which enables enhanced rendering and new visualization techniques. Interactivity is of great importance for data exploration and analysis in order to gain insight into large-scale data. Increasingly large data sizes are pushing the limits of brute-force rasterization algorithms present in the most widely-used visualization software. Interactive ray tracing presents an alternative rendering solution which scales well on multicore shared memory machines and multinode distributed systems while scaling with increasing geometry counts through logarithmic acceleration structure traversals. Ray tracing within existing tools also provides enhanced rendering options over current implementations, giving users additional insight from better depth cues while also enabling publication-quality rendering and new models of visualization such as replicating photographic visualization techniques.