Scalable parallel direct volume rendering for nonrectilinear computational grids
Judith Challinger · 1993
Various approaches to the problem of parallelizing direct volume rendering algorithms are discussed, and a scalable approach to parallel direct volume rendering of nonrectilinear computational grids is presented. The algorithm is general enough to handle non-convex grids and cells, grids with voids, grids constructed from multiple grids (multi-block grids), and embedded geometrical primitives. The algorithm is designed for a highly parallel MIMD architecture which features both local memory and shared memory with non-uniform memory access times. It has been implemented on a BBN TC2000 and benchmarked on several datasets. An analysis of the speedup and efficiency of the algorithm is given and any sources of inefficiency are identified. The trade-off between load balancing requirements and the loss of coherence due to the image-space task decomposition is examined. A variation of the algorithm which provides fast image updates for a changing transfer function is also presented. A distributed approach to controlling the execution of the volume render is used and the graphical user interface designed for this purpose is briefly described.