A divided-screenwise hierarchical compositing for sort-last parallel volume rendering

In this work, to render at least 512/sup 3/ voxel volumes in real-time, we have developed a sort-last parallel volume rendering method for distributed memory multiprocessors. Our sort-last method consists of two methods, Hsu's (1993) segmented ray casting and our divided-screenwise hierarchical (DSH) compositing, in which each processor produces a subimage and merges all the produced subimages into the final image. This paper describes the DSH method, which aims at achieving high performance compositing on a large number of processors. Our implementation on a 64-node PC cluster can composite a 512/sup 2/ pixel image about twice as fast as an existing method, the binary-swap method, so that can render a 512/spl times/512/spl times/224 voxel volume at approximately eight frames per second (fps).