Parallel Volume Rendering with Early Ray Termination for Visualizing Large-Scale Datasets
This paper presents an efficient parallel algorithm for volume rendering of large-scale datasets. Our algorithm focuses on an optimization technique, namely early ray termination (ERT), which aims to reduce the amount of computation by avoiding enumeration of invisible voxels in the visualizing volume. The novelty of the algorithm is that it incorporates this technique into a distributed volume rendering system with global reduction of the computational amount. The algorithm also is capable of statically balancing the processor workloads. The experimental results show that our algorithm with global ERT further achieves the maximum reduction of 33% compared to an earlier algorithm with local ERT. As a result, our load-balanced algorithm reduces the execution time to at least 66%, not only for dense objects but also for transparent objects.
Unable to display preview. Download preview PDF.
- 4.Nukata, M., Konishi, M., Goshima, M., Nakashima, Y., Tomita, S.: A volume rendering algorithm for maximum spatial locality of reference. IPSJ Trans. Advanced Computing Systems 44, 137–146 (2003) (in Japanese)Google Scholar
- 5.Hsu, W.M.: Segmented ray casting for data parallel volume rendering. In: Proc. 1st Parallel Rendering Symp (PRS 1993), pp. 7–14 (1993)Google Scholar
- 6.Ma, K.L., Painter, J.S., Hansen, C.D., Krogh, M.F.: Parallel volume rendering using binary-swap compositing. IEEE Computer Graphics and Applications 14, 59–68 (1994)Google Scholar
- 8.Gao, J., Huang, J., Shen, H.W., Kohl, J.A.: Visibility culling using plenoptic opacity functions for large volume visualization. In: Proc. 14th IEEE Visualization Conf. (VIS 2003), pp. 341–348 (2003)Google Scholar
- 9.Lee, C.H., Park, K.H.: Fast volume rendering using adaptive block subdivision. In: Proc. 5th Pacific Conf. Computer Graphics and Applications (PG 1997), pp. 148–158 (1997)Google Scholar
- 10.O’Carroll, F., Tezuka, H., Hori, A., Ishikawa, Y.: The design and implementation of zero copy MPI using commodity hardware with a high performance network. In: Proc. 12th ACM Int’l Conf. Supercomputing (ICS 1998), pp. 243–250 (1998)Google Scholar