high performance computing on graphics processing units: hgpu.org

More and more 3D terrain information is converted and stored in a digital format, and they have been used as a primary data source for a variety of spatial analyses in various application domains. One unsolved issue is that 3D spatial analyses (e.g., 3D ray tracing) often require huge computing cost in terms of processing time and hardware. This paper presents a new technique for the fast 3D visibility analysis to find relevant building surfaces for rapid wave propagation analysis. At the begging of this paper, a set of algorithm for fast 3D spatial indexing algorithm is described. A graphic process unit (GPU) that assists distributed computing process is also introduced in this paper as well. The performance of the proposed algorithm was tested using a set of 3D building models in Daejeon, Korea. Preliminary test results shows that the proposed method can be used for complex 3D spatial analysis with datasets of large volumes and in wide area.