GPU and CPU cooperation parallel visualisation for large seismic data

Abstract

A parallel visualisation technique for large seismic data based on a graphics processing unit (GPU) and CPU cooperation is presented. The 3D texture mapping capability commonly available in modern GPUs is employed as a core rendering engine and advantage is taken of combinations of GPU-based parallel volume visualisation and visibility tests to accelerate the whole rendering process. Experimental results show that the method achieves real-time performance for very large seismic data (10GB) on standard PC hardware.