Data Local Iterative Methods For The Efficient Solution of Partial Differential Equations

A cooperation
between

and
.

Funded by
.

Talks

2007:

• Ulrich Rüde. Challenges and potentials of emerging multicore architectures. 3rd Joint HLRB and KONWIHR Result and ReviewingWorkshop, LRZ, Garching, December 2007.
  Slides (pdf)
• Josef Weidendorfer. Understanding Memory Access Bottlenecks on Multicore. Talk at ParCo 2007, September 2007.
  Slides (pdf)
• Jan Götz and Markus Stürmer. Simulation of Blood Flow in the Human Brain Using the Cell Processor. The Fourth International Conference for Mesoscopic Methods in Engineering and Science (ICMMES), München, July 2007.
  Slides (pdf)
• Josef Weidendorfer. Single Processor Performance: Analysis with Callgrind / KCachegrind. Talk at Multicore-Training, LRZ, Garching, July 2007.
  Slides (pdf)
• Josef Weidendorfer. Single Processor Performance: Analysis with Callgrind / KCachegrind. Talk at 1st Parallel Tools Workshop, HLRS, Stuttgart, July 2007.
  Slides (pdf)
• Markus Stürmer. Optimizing Fluid Simulation and other scientific applications on the Cell. SFB 716, Stuttgart, June 2007.
  Slides (pdf)
• Markus Stürmer. Fluid Simulation using the Lattice Boltzmann Method on the Cell Processor. Zentralinstitut für Angewandte Mathematik des Forschungszentrums Jülich, Jülich, April 2007.
  Slides (pdf)
• Ulrich Rüde, Benjamin Bergen, Christoph Freundl, and Tobias Gradl. Massively Parallel Multigrid for Finite Elements. 13th Copper Mountain Conference on Multigrid Methods, Copper Mountain, USA, 20.03.2007.
  Slides (pdf)
• Markus Stürmer, Harald Köstler, and Ulrich Rüde. A fast full multigrid solver for applications in image processing. 13th Copper Mountain Conference on Multigrid Methods, Copper Mountain, USA, March 2007.
  Slides (pdf)

2006:

• Markus Stürmer, Jan Treibig, and Ulrich Rüde. Optimizing a 3D Multigrid for the IA-64 Architecture. ASIM 2006 - 19th Symposium Simulationstechnique, ASIM, Hannover, September 2006.
  Slides (pdf)
• Aditya Nitsure, Klaus Iglberger, Ulrich Rüde, Christian Feichtinger, Gerhard Wellein, and Georg Hager. Optimization of Cache Oblivious Lattice Boltzmann Method in 2D and 3D. ASIM 2006 - 19th Symposium Simulationstechnique, ASIM, Hannover, September 2006.
  gzipped Slides (Powerpoint) and movies
• Gerhard Wellein, Thomas Zeiser, Georg Hager, Aditya Nitsure, Klaus Iglberger, and Ulrich Rüde. Introducing a Cache-Oblivious Blocking Approach for the Lattice Boltzmann Method. ICMMES 06, Hampton (VA), USA, July 2006.
  Slides (pda)
• Jan Treibig. Why is the Performance Productivity Poor on Modern Computer Architectures. Seminar 06071 "Architectures and Algorithms for Petascale Computing", Dagstuhl, February 2006.
  Slides (pdf)

2005:

• Josef Weidendorfer. Simulation by Runtime Instrumentation and Profile Visualization. Demonstration on Workshop on Automatic Performance Analysis (WAPA 2005), Dagstuhl, Germany, December 2005.
• Ulrich Rüde and Jan Treibig . Optimizing Numerical Simulation Kernels for Current Instruction and Memory Architectures. ScalPerf 05, Bertinoro, October 2005.
  Slides (pdf)
• Jan Treibig. Performance analysis of the Lattice Boltzmann Method on x86-64 Architectures. ASIM05, IMMD10, Erlangen, September 2005.
  Slides (pdf)
• Stefan Donath. Cache-Optimized Implementations of the Lattice Boltzmann Methods. ASIM05, IMMD10, Erlangen, September 2005.
  Slides (pdf)
• Jochen Härdtlein. Blocking Techniques with Fast Expression Templates. POOSC-Workshop, 19th European Conference on Object-Oriented Programming, University of Glasgow, Glasgow, July 2005.
  Slides (pdf)
• Georg Hager, Thomas Zeiser, Jan Treibig, and Gerhard Wellein. Optimizing performance on modern HPC systems: learning from simple kernel benchmarks. The 2nd Russian-German Advanced Research Workshop on Computational Science and High Performance Computing, German-Russian Center for Computational Technologies and High Performance Computing, Stuttgart, March 2005.
  Slides (pdf)
• Klaus Iglberger. Cache Optimization for the Lattice Boltzmann Method in 3D. CSE Conference 2005, SIAM, Orlando, 14.02.2005, February 2005.
  Slides (pdf)

2004:

• Ulrich Rüde, Markus Kowarschik, Arndt Bode, and Josef Weidendorfer. Performance Prediction, Analysis, and Optimization of Numerical Methods on Cache-Based Computer Architectures. Half-Day Tutorial at the 31st Annual International Symposium on Computer Architecture (ISCA 2004), Munich, Germany, June 2004.
• Ulrich Rüde. Cache-Optimizations for Numerical Algorithms. Dagstuhl- Seminar on Cache-Oblivious and Cache-Aware Algorithms, July 2004.
• Markus Kowarschik, Iris Christadler, and Ulrich Rüde. Patch-Adaptive Relaxation in Multigrid Algorithms. PARA´04 State-of-the-Art in Scientific Computing, Technical University of Denmark, Lyngby, Copenhagen, Denmark, June 2004. Slides (pdf)
• Ulrich Rüde et al. Experiences with the design of cache-aware numerical algorithms, ScalPerf04, Bertinoro, Italy, September 2004.
  Slides (ppt.gz)
• Ulrich Rüde et al. Performance-/Cache-Optimierungen in der numerischen Simulation am Beispiel von Lattice-Boltzmann-Verfahren, ASIM-Workshop der GI, Siegen, Germany, March 2004.
  Slides (pdf)
• Markus Kowarschik. Cache Performance Optimizations for Iterative Linear Solvers and the Lattice Boltzmann Method, TU Wien, Vienna, Austria, May 2004.
  Slides (pdf)
• Josef Weidendorfer, Markus Kowarschik, Carsten Trinitis. A Tool Suite for Simulation Based Analysis of Memory Access Behavior. International Conference on Computational Science (ICCS 2004), Krakov, Poland, June 2004.
  Slides (pdf)
• Josef Weidendorfer, Carsten Trinitis. Cache Optimization for Iterative Numerical Codes Aware of Hardware Prefetching. International Conference on Applied Parallel Computing (PARA'04), Copenhagen, Denmark, June 2004.
  Slides (pdf)

2003:

• Markus Kowarschik. Hardware-Oriented Locality Optimizations for Iterative Methods. Oberseminar Wissenschaftliches Rechnen, TU München, Germany, Januar 2003.
  Slides (pdf)
• Ulrich Rüde, Markus Kowarschik. Locality-Optimized Iterative Methods. SIAM Conference on Computational Science and Engineering, San Diego, CA, USA, February 2003 (Organization of a minisymposium on "Locality in Scientific Applications").
  Slides (pdf)
• Markus Kowarschik, Ulrich Rüde. Cache Performance Optimizations for Multigrid Computations. Seminar on Challenges in High Performance Simulations for Science and Engineering, Schloss Dagstuhl, Germany, March 2003.
  Slides (pdf)
• Thomas Pohl. Performance-Optimized Lattice Boltzmann. Seminar on Challenges in High Performance Simulations for Science and Engineering, Schloss Dagstuhl, Germany, March 2003. Joint work between our project and the FreeWiHR project.
  Slides (pdf)

2002:

• Markus Kowarschik, Christian Weiß. Cache-Aware Numerical Computations. GI Dagstuhl Forschungsseminar, Schloss Dagstuhl, Germany, March 2002.
  Slides (pdf)
• Ulrich Rüde, Markus Kowarschik, Nils Thürey, Christian Weiß. Cache Optimizations for Multigrid Codes. International Conference on Computational Science (ICCS 2002), Amsterdam, The Netherlands, April 2002.
  Slides (pdf)
• Markus Kowarschik, Ulrich Rüde, Nils Thürey, Christian Weiß. Performance Optimization of Multigrid Codes on Hierarchical Memory Architectures. International Conference on Applied Parallel Computing (PARA'02), Espoo, Finland, June 2002.
  Slides (pdf)

2001:

• Markus Kowarschik, Ulrich Rüde, Christian Weiß, W. Karl. Cache-friendly Multigrid Codes. GAMM 2001 Annual Meeting, ETH Zürich, Switzerland, February 2001 (Organization of a minisymposium on "Hardware-Oriented Numerics").
  Slides (ps.gz)
• Markus Kowarschik. Enhancing the Cache Performance of Multigrid Codes on Structured Grids. Copper Mountain Conference on Multigrid Methods, Copper Mountain, Colorado, USA, April 2001.
  Slides (ps.gz)
• Markus Kowarschik, Christian Weiß. DiMEPACK: A Cache-Optimized Multigrid Library. International Conference on Parallel and Distributed Processing Techniques and Applications (PDPTA 2001), Las Vegas, Nevada, USA, June, 2001.
  Slides (ps.gz)

1999:

• Thomas Pröll. Hardwaresimulation mit SimOS. Internal talk, TU München, Munich, February 1999.
  Slides (ps.gz)
• Elfriede Brem. Wisconsin Architecural Research Tool Set: An Overview. Internal Talk, TU München, Munich, February 1999.
  Slides (ps.gz)
• Markus Kowarschik, Christian Weiß, Jonathan Hu, Craig C. Douglas, Ulrich Rüde, Wolfgang Karl. How much can be gained by Cache Aware Programming of Multigrid Methods? 9th Copper Mountain Conference on Multigrid Methods, Colorado, April 1999.
  Slides (ps.gz)
• Markus Kowarschik, Christian Weiß, Jonathan Hu, Craig C. Douglas, Ulrich Rüde, and Wolfgang Karl. Cache Optimization Techniques for Numerical Computations. SPEC Benchmark Workshop, Wuppertal, Paderborn, September 1999.
  Slides (ps.gz)
• Craig C. Douglas, Jonathan Hu, Markus Kowarschik, Ulrich Rüde, Christian Weiß. Fixed and Adaptive Cache Aware Algorithms for Multigrid Methods. Sixth European Multigrid Conference, Gent, Belgium, September 1999.
  MS Powerpoint Slides (ppt.gz)
• Ulrich Rüde, Christian Weiß, Markus Kowarschik, Wolfgang Karl. Memory Characteristics of Iterative Methods. Supercomputing Conference, Portland, Oregon, November 1999.
  Slides (ps.gz)

1998:

• Christian Weiß, Hermann Hellwagner, Linda Stals, and Ulrich Rüde. Data Locality Optimizations to Improve The Effiziency of Multigrid Methods. GAMM Seminar'98, Christian-Albrechts-Universität, Kiel, January 1998.
  Slides (ps.gz)
• Ulrich Rüde. Technological Trends and their Impact on the Future of Supercomputing. International FORTWIHR Conference on HPSEC, München, March 1998.
  
• Christian Weiß. Datenlokalitätsoptimierung für Architekturen mit tiefen Speicherhierarchien. Hauptseminar SS98, Technische Universität München, Munich, July 1998.
  Slides (ps.gz)
• Markus Kowarschik, Ulrich Rüde, and Christian Weiß. Cache-aware Multigrid for 2- and 3-dimensional Elliptic Equations. 10th International Workshop on Multigrid Methods, Bonn, October 1998.
  Slides (ps.gz)

1997:

• Ulrich Rüde. Iterative Algorithms on High Performance Architectures. Copper Mountain Conference on Multigrid Methods 1997, Colorado, April 1997.
  Slides (ps.gz)
• Ulrich Rüde. Iterative Algorithms on High Performance Architectures. EuroPar97 Conference, Passau, August 1997.
  
• Linda Stals, Ulrich Rüde, Christian Weiß, and Hermann Hellwagner. Data Local Iterative Methods For The Efficient Solution of Partial Differential Equation. Tenth International Conference on Domain Decomposition Methods, Colorado, August 1997.
  Slides (tar.gz)

• Ulrich Rüde, Linda Stals, Christian Weiß, and Hermann Hellwagner. Data Local Iterative Methods for the Efficient Solution of Partial Differential Equations. Eighth Biennial Computational Techniques and Applications Conference, Adelaide, Australia, September 1997.