In this paper we present a parallelized particle-based volume rendering (PBVR) system to obtain super linear speedup, i.e. the speedup with N processors is greater than N, on a distributed computing system. PBVR is a technique of volume rendering which does not need any sorting therefore it is suitable for parallel execution. We implemented the parallel version by using Open CABIN middleware and measured the performance. We confirmed that the parallelized PBVR is 1.5 to 8 times faster than the serial version.
展开▼
