HR: 10:50h
AN: IN42A-03    [Abstracts]
TI: gLucifer: Next-generation visualization framework for high-perfomance computational geodynamic models
AU: Duboz, C
EM: Cecile.Duboz@sci.monash.edu.au
AU: * Stegman, D
EM: dave.stegman@sci.monash.edu.au
AF: Monash Cluster Computing, School of Mathematical Sciences, Monash University, Clayton, VIC 3121 Australia
AU: Moresi, L
EM: Louis.Moresi@sci.monash.edu.au
AF: Monash Cluster Computing, School of Mathematical Sciences, Monash University, Clayton, VIC 3121 Australia
AU: Turnbull, R
EM: robert.turnbull@sci.monash.edu.au
AF: Monash Cluster Computing, School of Mathematical Sciences, Monash University, Clayton, VIC 3121 Australia
AU: Lo, A
EM: alan@vpac.org
AF: Victorian Partnership For Advanced Computing, Bldg 91, 110 Victoria Street, Carlton South, VIC 3053 Australia
AU: Quenette, S
EM: steve@vpac.org
AF: Victorian Partnership For Advanced Computing, Bldg 91, 110 Victoria Street, Carlton South, VIC 3053 Australia
AB: High performance computing provides unprecedented capabilities to produce higher resolution 4-D models in a fraction of the time. Thus, the need exists for a new generation of visualization systems able to maintain parity with the enormous volume of data generated from state-of-the-art numerical models. However, this volume of data is too large to be written to disk, even if only temporarily, as it introduces a significant performance bottleneck and therefore requires an innovation to circumvene the assumption that most existing visualization software make at a quite fundamental level, which is that model data is input from a dump file. This issue, in conjunction with the fact that available packages do not scale well for plotting very large numbers of specialized particles, necessitates the creation of a new visualization system that meets the needs of large-scale geodynamic modeling. We have successfully developed such a system, gLucifer, using a software framework approach that allows efficient re-use of our efforts in other areas of research. gLucifer is capable of producing movies of a 4-D data set ``on the fly'' (simultaneously with running the parallel scientific application) and sampled at the finest spatial and temporal scales (the highest resolution every timestep), without creating a performance bottleneck. By eliminating most of the human effort involved in visualizing results through post-processing, gLucifer reconnects the scientist to the numerical experiment as it unfolds in real time. gLucifer further provides a novel type of interactivity, termed iterative visualization, that exploits the ever-increasing availability of computational resources. Through the use of multiple windows with multiple viewports, generating custom fields ``on the fly'' with variables and operators, and fully distributed rendering and scene compositing, data sets that were previously very difficult to even manage may be efficiently explored and interrogated. Future developments include the ability to post-process fully rendered movies and harness the potential of remote visualization enabled with grid computing.
DE: 0525 Data management
DE: 0530 Data presentation and visualization
DE: 0550 Model verification and validation
DE: 5430 Interiors (8147)
DE: 5724 Interiors (8147)
SC: Earth and Space Science Informatics [IN]
MN: Fall Meeting 2005