Distributed Volume Rendering of Very Large Data on High-Resolution Scalable Displays
A volume visualization of a Purkinje neuron on the LambdaTable - Lance Long, EVL
Developers: Nicholas Schwarz
Scientific instruments increasing collect higher spatial resolution samples providing scientists with larger volumetric datasets.
For example, bioscientists at the National Center for Microscopy and Imaging Research (NCMIR) at the University of California, San Diego (UCSD) regularly collect high-resolution data from high-powered multi-photon microscopes. Geologists at the University of Minnesota (UMN) often collect high-resolution CT scans of geologic samples.
Scalable high-resolution tiled-displays allow scientists to visualize this large data at or closer to its native resolution and are seeing an increasing rate of adoption by the scientific community. Currently, no solution exists that allows scientists to visualize this large data and view it at its full spatial resolution.
This project provides a volume visualization solution that allows scientists to render very large volumetric datasets on scalable high-resolution displays. It uses a methodology that employs a multi-resolution octree, an image-order data distribution method, a distributed shared-memory data management system, a multi-level cache, and hardware accelerated rendering techniques to produce a solution that is scalable in terms of input data size and output resolution.
An analytical cost model validated by experimental results predicts the system’s behavior. The methodology’s usefulness is demonstrated with a number of domain specific datasets.
Date: January 1, 2006 - December 31, 2007