Program
Automated Visualization: the Future of In Situ Processing?
Supercomputing trends, in particular I/O rates, are forcing simulation codes away from the traditional model of storing data to disk and then visualizing post hoc, and towards an in situ model where data is processed as it is generated. While in situ processing solves many problems, it also creates new problems. In this presentation, I will focus on one of these problems, specifically that in situ processing frequently has no human in the loop, and so the visualization process must be automated. In cases where the desired visualization algorithms and their settings are known a priori, the automation process is not difficult. But in cases where the algorithms and their settings are unknown, it will be up to the visualization program to decide how to proceed. This new direction can tap into a decade of previous research by our community on automating visualizations, i.e., choosing the best camera locations, transfer functions, etc., by using an algorithm rather than a human directing the process. In this talk, I will summarize this decade of research, with a special focus on my work.
Ivan Viola is an Associate Professor at King Abdullah University of Science and Technology (KAUST), Saudi Arabia. In his research, Viola investigates methods that automate the visualization design process. Basal new technologies he then applies to various application domains, such as ultrasound diagnostics, geology, or sub-millimeter biology. Viola has graduated from TU Wien, Austria in 2005 and moved for postdoctoral fellowship to the University of Bergen, Norway, where he was gradually promoted to the rank of Professor. In 2013 he has been awarded a Vienna Science and Technology Fund grant to establish his own research group back at TU Wien, Austria. After several years of research on visualization and computer graphics techniques for multiscale biological structures in Vienna, Viola has co-founded the startup company called Nanographics, to commercialize nanovisualization technologies. Currently, Viola has relocated to KAUST for taking advantage of generous research conditions to multiply his investigative efforts, for delivering technologies that make visual, in-silico life possible.
Monday, June 3rd
Location: Room Porto
Registration: 8:30-9:00
Session 1: 9:00 - 10:40
Opening
Dynamic I/O Budget Reallocation For In Situ Wavelet Compression
- Nicole J Marsaglia: CDUX, University of Oregon, Eugene, Oregon, United States
- Shaomeng Li: National Center for Atmospheric Research, Boulder, Colorado, United States
- Kristi Belcher: University of Oregon, Eugene, Oregon, United States
- Matthew Larsen: Lawrence Livermore National Laboratory, Livermore, California, United States
- Hank Childs: University of Oregon, Eugene, Oregon, United States
Fast Mesh Validation in Combustion Simulations through In-Situ Visualization
- Sergei Shudler: Argonne National Laboratory, Lemont, Illinois, United States
- Nicola Ferrier: Argonne National Laboratory, Chicago, Illinois, United States
- Joseph Insley: Argonne National Laboratory, Lemont, Illinois, United States
- Michael E. Papka: Argonne National Laboratory, Chicago, Illinois, United States
- Saumil Patel: Argonne National Laboratory, Lemont, Illinois, United States
- Silvio Rizzi: Argonne National Laboratory, Chicago, Illinois, United States
Scalable Parallel Flow Visualization Using 3D Line Integral Convolution for Large Scale Unstructured Simulation Data
- Yangguang Liao: Computer Science, University of California, Davis, Davis, California, United States
- Hiroaki Matsui: Computational Infrastructure for Geodynamics, University of California, Davis, Davis, California, United States
- Oliver Kreylos: W.M. Keck Center for Active Visualization in the Earth Sciences (KeckCAVES), University of California, Davis, Davis, California, United States
- Louise H. Kellogg: Earth and Planetary Sciences, University of California, Davis, Davis, California, United States
Task-based Augmented Reeb Graphs with Dynamic ST-Trees
- Charles Gueunet: Kitware SAS, Lyon, France
- Dr. Pierre Fortin: LIP6, Sorbonne Université, Paris, France
- Julien Jomier: Kitware, Lyon, France
- Julien Tierny: CNRS, Paris, France
Session 2: 11:10 - 12:50
Parallel XPBD Simulation of Modified Morse Potential - an Alternative Spring Model
- Dr. Ozan Cetinaslan: Instituto de Telecomunicações & Faculdade de Ciências, Universidade do Porto, Porto, Portugal
Real-time particle-based snow simulation on the GPU
- Dr Prashant Goswami: Computer Science, BTH, Karlskrona, Sweden
- Christian Markowicz: Computer Science, BTH, Karlskrona, Sweden
- Ali Hassan: Computer Science, BTH, Karlskrona, Sweden
Hybrid Online Autotuning for Parallel Ray Tracing
- Killian Herveau: Karlsruhe Institute of Technology, Karlsruhe, Germany
- Philip Pfaffe: Karlsruhe Institute of Technology, Karlsruhe, Germany
- Martin Peter Tillmann: Karlsruhe Institute of Technology, Karlsruhe, Germany
- Walter F. Tichy: Karlsruhe Institute of Technology, Karlsruhe, Germany
- Carsten Dachsbacher: Karlsruhe Institute of Technology, Karlsruhe, Germany
Screen Partitioning Load Balancing for Parallel Rendering on a Multi-GPU Multi-Display Workstation
- Yangzi Dong: University of Alabama in Huntsville, Huntsville, Alabama, United States
- Chao Peng: University of Alabama in Huntsville, Huntsville, Alabama, United States
Session 3: 14:10 - 15:50
Efficient Point Merge Using Data Parallel Techniques
- Abhishek Yenpure: University of Oregon, Eugene, Oregon, United States
- Hank Childs: University of Oregon, Eugene, Oregon, United States
- Kenneth Moreland: Sandia National Laboratories, Albuquerque, New Mexico, United States
Keynote
Session 4: 16:20 - 18:00
Hybrid Remote Visualization in Immersive Virtual Environments with Vistle
- Martin Aumüller: HLRS, University of Stuttgart, Stuttgart, Germany
Statistical Analysis of Parallel Data Uploading using OpenGL
- Markus Wiedemann: Leibniz Supercomputing Centre, Bavarian Academy of Sciences and Humanities, Garching near Munich, Germany
- Dieter Kranzlmüller: MNM-Team, Ludwig-Maximilians-Universität München, Munich, Germany
An Interpolation Scheme for VDVP Lagrangian Basis Flows
- Sudhanshu Sane: Computer Science, University of Oregon, Eugene, Oregon, United States
- Hank Childs: University of Oregon, Eugene, Oregon, United States
- Dr. Roxana Bujack: Los Alamos National Laboratory, Los Alamos, New Mexico, United States
Closing
