Parallel volume rendering on virtual machines

A volume rendering algorithm for parallel virtual machines is presented. To reduce the communication cost, and to guarantee the locality of subtasks, the volume data is divided and organized as a series of slices. By maintaining and employing a database of the performance index, the task subdivision algorithm produces an acceptable load balancing. A binary asynchronous method is introduced to merge all partial images in O(nlogn) time. Our algorithm was implemented, exploiting the classical client/server paradigm. Scalabilities of both the task size and the number of hosts are tested and experimental results are demonstrated and analyzed.