Markov Prefetching in Multi-Block Particle Tracing

Marc Woltera; Andreas Gerndta; Torsten Kuhlena; Christian Bischofb    a Virtual Reality Group. RWTH Aachen Universityb Center for Computing and Communication, RWTH Aachen University

Visualization of flow phenomena in large, time varying data sets demands extensive graphical, computational and memory resources. Recent research in the field of parallel post-processing algorithms and out-of-core data management shows that a good utilization of given resources together with a considerable speed-up may be achieved. One of the main problems is the I/O bottleneck commonly alleviated by data caching and overlapping I/O with computation. Standard prefetching techniques suffice for predictable data demands ...

Get Parallel Computational Fluid Dynamics 2006 now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.