A rotate-tiling image composition method for parallel volume rendering on distributed memory multicomputers

Abstract

The binary-swap and the parallel-pipelined methods are two popular image composition methods for volume rendering on distributed memory multicomputers. However, these methods either restrict the number of processors to a power of two or require many steps to transform image data that results in high communication overheads. In this paper, we present an efficient image composition method, the rotate-tiling (RT), for parallel volume rendering on distributed memory multicomputers. The RT method can fully utilize all available processors and minimize the communication overheads. In addition, we provide data compression method, the template run-length encoding (TRLE), to further reduce the communication data size. To evaluate the performance of the RT method, we compare the proposed method with the binary-swap method and the parallel-pipelined method. Both theoretical analysis and experimental test are conducted. In the theoretical analysis, we analyze the best performance bound of the RT method in terms of the startup time, the data transmission time, the number of processors, and the number of initial block of a sub-image. In the experimental test, we have implemented these three methods on an SP2 parallel machine. Three volume datasets are used as test samples. The experimental results show that our method outperforms the binary-swap and the parallel-pipelined methods for all test samples and match the results analyzed in the theoretical analysis. For the TRLE method, the experimental results show that the TRLE method can further reduce the composition time for these three methods.