Abstract

This paper presents a distributed architecture for the visualization of large steady and transient data-sets. Because one of the goals is to not have to store away the 10s to 100s of Gigabytes necessary to perform a post-processing animation of the data, this system must be able to run concurrently with the task producing the data. This requirement forces portions of the visualization system to run on the same hardware where the data is generated. What follows is a discussion of current computer architectures used for CFD calculations and how these systems affect the model selected for the visualization software. Also data and data movement considerations must be an integral part of any project that requires the distribution of a compute task. 1 Introduction Scientific visualization is an area that has grown greatly in size and importance in the last five years. This growth has been driven by a number of factors. One is the change in the types of calculations that are being done by Co...

Citations

128	A users’ guide to PVM parallel virtual machine - Beguelin, Dongarra, et al.
28	HINKER P.: Massively parallel isosurface extraction - HANSEN
20	M.: VISUAL3 Interactive Unsteady Unstructured 3D Visualization - HAIMES, GILES - 1991
15	Visualization in computational fluid dynamics: a case study - Haimes, Darmofal - 1991
11	A Software Model for Visualization of Time Dependent 3-D Computational Fluid Dynamics Results - Globus - 1992
6	Distributed Interactive Graphics Applications in Computational Fluid Dynamics - ROGERS, BUNING, et al. - 1987
1	Visualization of Time-dependent Flow Fields. submitted to - Lane - 1993
1	Implementing a Distributed Process between Workstation and Supercomputer - Grimsrud, Lorig - 1989