Issue No.08 - August (2011 vol.17)
Marco Ament , Universität Stuttgart, Stuttgart
Daniel Weiskopf , Universität Stuttgart, Stuttgart
Brendan Moloney , Oregon Health Science University, Portland
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/TVCG.2010.116
Sort-first distributions have been studied and used far less than sort-last distributions for parallel volume rendering, especially when the data are too large to be replicated fully. We demonstrate that sort-first distributions are not only a viable method of performing data-scalable parallel volume rendering, but more importantly they allow for a range of rendering algorithms and techniques that are not efficient with sort-last distributions. Several of these algorithms are discussed and two of them are implemented in a parallel environment: a new improved variant of early ray termination to speed up rendering when volumetric occlusion occurs and a volumetric shadowing technique that produces more realistic and informative images based on half angle slicing. Improved methods of distributing the computation of the load balancing and loading portions of a subdivided data set are also presented. Our detailed test results for a typical GPU cluster with distributed memory show that our sort-first rendering algorithm outperforms sort-last rendering in many scenarios.
Volume rendering, sort-first parallelization, visualization, dynamic load balancing, early ray termination, shadow, ray coherence.
Marco Ament, Daniel Weiskopf, Brendan Moloney, "Sort-First Parallel Volume Rendering", IEEE Transactions on Visualization & Computer Graphics, vol.17, no. 8, pp. 1164-1177, August 2011, doi:10.1109/TVCG.2010.116