In this talk, I will present two algorithms for volume rendering large unstructured data sets that we are developing for the VIEWS Visualization Program. The first is based on the incremental slicing algorithm of Yagel, Reed, Law, Shih and Shareef. It is being parallelized for use with distributed data on the Stanford-TriLabs Visualization Supercomputer (STVC), a scalable cluster of graphics-enabled PC's connected by multiple high speed networks. I will also describe the STVC architecture and the motivation for its design.

The second algorithm is for SMP parallel rendering of monolithic data sets using multiple graphics engines, either IR pipes or graphic cards. The current implementation is for a SGI Onyx2 with 8 IR pipes and 48 cpus.

Snacks will be provided.

See Conundrum Talks for more information about this series.