Dynamic Sharing of Large-Scale Visualization

Driven by needs of SciDAC applications for large user groups to collaborate across geographical distances, visualization must now also serve as an effective means to share concrete data as well as abstract ideas over the Internet. Yet there is simply no expeditious and practical way for users collaborating in this wide area to share large visualizations in a dynamic fashion. Current users cannot dynamically interact with and tweak a visualization to answer questions formed spontaenously in his or her mind. We propose to use free, unscheduled, and unreserved resources on remote networked computers in an innovative way to visualize large datasets. We have shown that fault tolerance as well as scalability can be achieved. Our prototype system has been demonstrated on SC'05 and SC'06. More details of our work are available in our article in IEEE Computer Graphics and Application, Vol. 27, No. 1, 2007.

In the above image, we illustrate runtime usage of our system
with TSI as the sample application.

Credits: The TSI data was provided by John Blondin and
Anthony Mezzacappa under the auspices of the DOE SciDAC
Terascale Supernova Initiative.