A Scalable Infrastructure for Computational Steering
To fully harvest the potential of advanced computing technologies, the efficient and timely analysis of huge amounts of data is as essential as are front-line simulations. To optimize this process, one increasingly tries to put experts and their capabilities into the centre of the analysis and to support them by statistical tools whose complexity is mainly hidden. This requires new possibilities to interactively guide the exploration process by exploiting the humans’ perceptual and cognitive abilities. At the same time it has to be ensured that gained information can seamlessly be fed back into the simulation process to steer the simulation towards optimum performance. For such interoperability, human-computer interfaces, analytics algorithms, and software systems have to be refined accordingly.
The goal of this project is to design and prototype a scalable infrastructure for computational steering. It will be targeted for the computational engineering domain, which allows us to leverage existing cooperative developments as a starting point and to use real-world data that is representative in size, modality, and structure to what is available in other scientific areas like geology or biology. The infrastructure implements a processing pipeline ranging from scalable data processing workflows to interactive visualisation and human-computer interaction in virtual and augmented reality environments.
Our approach to handle the scalability issue we aim to cope with is to realize this pipeline on parallel visual computing architectures, including multi-core and multi-GPU systems. We intend to make adaptable steering facilities for explorative visual data analysis available across a range of facilities, from the KAUST Research Laboratories down to individual desktop settings. One key aspect will be the design of a generic infrastructure, parts of which can be used for computational steering and visual computing in essentially all KAUST Research Institutes.
- In Subproject Westermann the Closest Point Method for numerically approximating PDEs on general smooth surfaces for which a distortion-free parameterization does not exist or is difficult to compute has been extended towards the particular needs in computational steering. Also, n interactive high quality volume rendering method for particle simulation data has been developed. The method avoids the order-dependent resampling of particle quantities along the view rays by using a view-space discretization of the simulation domain.
- Subproject Rank introduced an integration framework for engineering applications that supports distributed computations as well as visualisation on-the-fly and enables a high degree of interactivity. Moreover, the problem of long communication delays was tackled in the case of huge data advent, which occur due to rigid coupling of simulation back-ends with visualisation front-ends and handicap a user in exploring intuitively the relation of cause and effect.
- Subproject Bungartz considers (visual) data exploration due to parameter variation. The simulation is treated as a function of various simulation parameters which is then numerically represented on a sparse grid. A first simple computational fluid dynamics (CFD) scenario, the so-called driven cavity, was used a reference simulation and good results in terms of accuracy but also interaction rates were obtained.
|Prof. Dr. Rüdiger Westermann (coordination)||Computer Graphics & Visualization|
|Prof. Bernd Brügge, Ph.D.||Applied Software Engineering|
|Prof. Dr. Hans-Joachim Bungartz||Scientific Computing in Computer Science|
|Prof. Claudia Czado, Ph.D.||Mathematical Statistics|
|Prof. Gudrun Klinker, Ph.D.||Augmented Reality|
|Prof. Dr. Ernst Rank||Computation in Engineering|
|Prof. Dr.-Ing. Wolfgang A. Wall||Computational Mechanics|
|Dr. Stefan Auer|
|Dr. Daniel Butnaru|
|Dr. Jovana Knežević|
|Dr. Hamman de Vaal|
Partners: NVIDIA, ATI, Intel
Poster (PDF) - Stefan Auer
Poster (PDF) - Daniel Butnaru
Poster (PDF) - Jovana Knežević
Poster (PDF) - Hamman de Vaal
- Michael Hamman de Vaal, Computational modeling, clinical comprehension and improvement of aortic manipulation, Dissertation, 2015
- Stefan Auer, Roland Fraedrich, Rüdiger Westermann, Efficient High-Quality Volume Rendering of SPH Data , IEEE Visualization 2010
- Stefan Auer, Rüdiger Westermann, Interactive Editing of GigaSample Terrain Fields , Computer Graphics Forum 31(2) (Proc. Eurographics 2012)
- S. Auer, C.B. Macdonald, M. Treib, J. Schneider, R. Westermann, Real-Time Fluid Effects on Surfaces using the Closest Point Method , Computer Graphics Forum 31: 6 (2012)
- Stefan Auer, Rüdiger Westermann, Direct Contouring of Implicit Closest Point Surfaces , Eurographics 2013 - Short Papers
- Stefan Auer, Rüdiger Westermann, An Embedding Method for Interactive Simulation on Dynamic Surfaces , Dissertation, 2013 (mediaTUM)
- Stefan Auer, Rüdiger Westermann, A Semi-Lagrangian Closest Point Method for Deforming Surfaces , Computer Graphics Forum 32 (7) (Proc. Pacific Graphics 2013)
- Jovana Knežević, Jérôme Frisch, Ralf-Peter Mundani, Ernst Rank, Interactive Computing Framework for Engineering Applications , INTERCOMP 2011
- Yang Li, Nitesh Narayan, Jonas Helming, Maximilian Koegel, A Domain Specific Requirements Model for Scientific Computing , ICSE 2011
- A. Murarasu, J. Weidendorfer, G. Buse, D. Butnaru and D. Pflüger, Compact Data Structure and Scalable Algorithms for the Sparse Grid Technique , PPoPP 2011
- Daniel Butnaru, Dirk Pflüger, Hans-Joachim Bungartz, Towards High-Dimensional Computational Steering of Precomputed Simulation Data using Sparse Grids , ICCS 2011
- D. Butnaru, G. Buse and D. Pflüger, A Parallel and Distributed Surrogate Model Implementation for Computational Steering ,
- Daniel Butnaru, Hans-Joachim Bungartz, Fast Insight into High-Dimensional Parametrized Simulation Data ,
- Daniel Butnaru, Computational Steering with Reduced Complexity , Dissertation, 2013 (mediaTUM)
- B. Peherstorfer, D. Butnaru, K. Willcox and H.-J. Bungartz, Localized Discrete Empirical Interpolation Method , SIAM Journal on Scientific Computing
- M. Hamman de Vaal, Michael W. Gee, Ulrich A. Stock, Wolfgang A. Wall , Complex FSI Models for Surgical Manipulations of the Human Aorta , FEF 2011
- Jovana Knežević, A High-Performance Computational Steering Framework for Engineering Applications , Dissertation, 2013 (mediaTUM)