Dr. rer. nat. Wolfgang Eckhardt

Address:

TU München

Institut für Informatik

Boltzmannstr. 3

85748 Garching b. München

Office:

MI 02.05.055

Email:

Phone:

(089) 289 18 632

Fax:

(089) 289 18 607

Office Hours:

by arrangement

Interests

• Algorithms and Datastructures for High Performance Computing
• Molecular Dynamics Simulations
• GPGPU-Programming

PRACE ISC Award 2013

Together with my co-authors, I received the PRACE ISC Award 2013 for our paper "591 TFLOPS Multi-Trillion Particles Simulation on SuperMUC", in which we describe our optimisations for the code Mardyn and perform the world's largest molecular dynamics simulation at a peak performance of 9.4 % on SuperMUC at LRZ.

Students

Finished Student Projects

Volume 2014

• A. Stephan: Entwicklung eines skalierbaren Checkpoint-Restart Schemas mit parallelem File-I/O [pdf] [BibTeX].
  Bachelor's thesis, Institut für Informatik, Technische Universität München, September 2014.
• M. Obersteiner: Parallel Cluster Detection in Nucleation Scenarios [pdf] [BibTeX].
  Bachelor's thesis, Institut für Informatik, Technische Universität München, September 2014.
• R. Hajda: Effiziente Vektorisierung des Dipolpotentials in parallelen Molekulardynamik-Simulationen [BibTeX].
  Bachelor's thesis, Institut für Informatik, Technische Universität München, February 2014.

Volume 2013

• W. Nicka: Keimdetektion in Gasen [BibTeX].
  Bachelor's thesis, Institut für Informatik, Technische Universität München, August 2013.
• J. Cremer: Effiziente Implementierung eines Zeitintegrationsalgorithmus mit CUDA [pdf] [BibTeX].
  Studienarbeit/SEP/IDP, Fakultät für Mathematik, Technische Universität München, January 2013.

Volume 2012

• M. Georgiev: Implementation of parallel Ewald-summation techniques [BibTeX].
  Master's thesis, Institut für Informatik, Technische Universität München, December 2012.
• T. Kovacevic: Aspects of Hybrid Molecular-Continuum Simulations for Complex Liquids [BibTeX].
  Master's thesis, Institut für Informatik, Technische Universität München, October 2012.
• A. C. Cavazos Sepulveda: Analysis of numerical precision requirements of Molecular Dynamics Simulations [BibTeX].
  Studienarbeit/SEP/IDP, Institut für Informatik, Technische Universität München, August 2012.
• J. Heckl: Effiziente Vektorisierung von Simulationen für starre mehrzentrige Molekülmodelle [pdf] [BibTeX].
  Bachelor's thesis, Institut für Informatik, Technische Universität München, July 2012.

Volume 2011

• P. Hoffmann: Ein Kopplungsansatz für hybride Lattice Boltzmann-Molekulardynamik-Simulationen mit MarDyn [pdf] [BibTeX].
  Master's thesis, Institut für Informatik, Technische Universität München, December 2011.
• A. Kirsch: Effiziente Implementierung von mehrzentrigen molekulardynamischen Potentialmodellen mit Cuda [pdf] [BibTeX].
  Studienarbeit/SEP/IDP, Institut für Informatik, Technische Universität München, October 2011.
• J. Weißl: Entwicklung eines effizienten Verfahrens zur Simulation langreichweitiger Molekül-Wechselwirkungen [BibTeX].
  Master's thesis, Institut für Informatik, Technische Universität München, October 2011.
• Y. Guo: Numerische Aspekte bei Molekulardynamik-Simulationen: Muster zur Speicheroptimierung der Linked-Cells Datenstruktur [pdf] [BibTeX].
  Studienarbeit/SEP/IDP, Institut für Informatik, Technische Universität München, January 2011.

Volume 2010

• K. Orend: Numerische Aspekte bei Molekulardynamik-Simulationen - Beschleunigung durch Graphikkarten [pdf] [BibTeX].
  Studienarbeit/SEP/IDP, Institut für Informatik, Technische Universität München, September 2010.
• J. Kummerow: Numerische Aspekte bei Molekulardynamik-Simulationen - Implementierung und Vergleich weiterer Zeitintegrationsverfahren [pdf] [BibTeX].
  Studienarbeit/SEP/IDP, Institut für Informatik, Technische Universität München, September 2010.

Teaching

PSE Molekulardynamik WS13

Tutorial: Scientific_Computing_II_-_Summer_13

PSE Molekulardynamik WS12

Tutorial: Introduction_to_Scientific_Computing_II_-_Summer_12

PSE Molekulardynamik WS11

Tutorial: Introduction_to_Scientific_Computing_II_-_Summer_11

PSE Molekulardynamik WS10
Masterpraktikum_Scientific_Computing_-_High_Performance_Computing_-_Winter_10
Tutorial: Algorithms_of_Scientific_Computing_II_-_Winter_10

Tutorial: Introduction_to_Scientific_Computing_II_-_Summer_10

Masterpraktikum_Scientific_Computing_-_High_Performance_Computing_-_Winter_09

Publications

Volume 2016

• H.-J. Bungartz, P. Neumann, N. Tchipev, W. Eckhardt and P. Jarmatz: Coupling 4 Molecular Dynamics Codes in a Massively Parallel Molecular-Continuum Fluid Dynamics Framework [BibTeX].
  In S. Wagner, A. Bode, H. Brüchle and M. Brehm (ed.), High Performance Computing in Science and Engineering Garching/Munich 2016, p. 156–157. Bayerische Akademie der Wissenschaften, München, July 2016.
• S. Becker, K. Stöbener, S. Werth, S. Eckelsbach, W. Eckhardt, A. Heinecke, N. Tchipev, H.-J. Bungartz, J. Vrabec and H. Hasse: High accuracy molecular dynamics simulation of fluids at interfaces [BibTeX].
  In S. Wagner, A. Bode, H. Brüchle and M. Brehm (ed.), High Performance Computing in Science and Engineering Garching/Munich 2016, p. 122–123. Bayerische Akademie der Wissenschaften, München, July 2016.

Volume 2015

• N. Tchipev, A. Wafai, C. W. Glass, W. Eckhardt, A. Heinecke, H.-J. Bungartz and P. Neumann: Optimized Force Calculation of Molecular Dynamics Simulations for the Intel Xeon Phi [BibTeX].
  In Euro-Par 2015: Parallel Processing Workshops, Volume 9523 of Lecture Notes in Computer Science, p. 774–785. Springer, Vienna, June 2015.
• A. Heinecke, W. Eckhardt, M. Horsch and H.-J. Bungartz: Supercomputing for Molecular-Dynamics Simulations: Handling Multi-Trillion Particles in Nanofluidics [BibTeX] of Springer Briefs in Computer Science. Springer, June 2015.

Volume 2014

• W. Eckhardt: Efficient HPC Implementations for Large-Scale Molecular Simulation in Process Engineering [BibTeX].
  Dissertation, Institut für Informatik, Technische Universität München, June 2014. Dissertation erhältlich im Verlag Dr. Hut unter ISBN 978-3-8439-1746-9 .
• W. Eckhardt: Efficient HPC Implementations for Large-Scale Molecular Simulation in Process Engineering [BibTeX]. Dr. Hut, München, June 2014.
• P. Neumann, W. Eckhardt and H.-J. Bungartz: Hybrid molecular-continuum methods: From prototypes to coupling software [BibTeX].
  In Computers and Mathematics with Applications, Volume 67 of Special issue ICMMES 2012, p. 272–281, January 2014.
• P. Neumann, W. Eckhardt and H.-J. Bungartz: A Radial Distribution Function-Based Open Boundary Force Model for Multi-Centered Molecules [] [BibTeX].
  In International Journal of Modern Physics C, Volume 25(6), 2014.
• H.-J. Bungartz, W. Eckhardt and A. Heinecke: Large-Scale Molecular Dynamics in Chemical Engineering [pdf] [BibTeX].
  In S. Wagner, A. Bode, H. Satzger and M. Brehm (ed.), High Performance Computing in Science and Engineering, p. 194–195. Leibniz-Rechenzentrum, Garching/Munich, 2014.

Volume 2013

• W. Eckhardt, A. Heinecke, W. Hoelzl and H.-J. Bungartz: Vectorization of Multi-Center, Highly-Parallel Rigid-Body Molecular Dynamics Simulations [pdf] [BibTeX].
  In Supercomputing 2013, The International Conference for High Performance Computing, Networking, Storage and Analysis.. IEEE, Denver, November 2013. Poster abstract..
• W. Eckhardt, A. Heinecke, R. Bader, M. Brehm, N. Hammer, H. Huber, H.-G. Kleinhenz, J. Vrabec, H. Hasse, M. Horsch, M. Bernreuther, C. Glass, C. Niethammer, A. Bode and H.-J. Bungartz: 591 TFLOPS Multi-Trillion Particles Simulation on SuperMUC [BibTeX].
  In International Supercomputing Conference (ISC) Proceedings 2013, Volume 7905 of Lecture Notes in Computer Science, p. 1–12. Springer, Heidelberg, Germany, June 2013.
• W. Eckhardt and A. Heinecke: SuperMUC boosts the largest molecular dynamics simulation by 4X in Number of Particles [BibTeX].
  In inside - Innovatives Supercomputing in Deutschland, Volume 11(1), p. 19–21, May 2013.

Volume 2012

• W. Eckhardt and T. Neckel: Memory-Efficient Implementation of a Rigid-Body Molecular Dynamics Simulation [html] [BibTeX].
  In Proceedings of the 11th International Symposium on Parallel and Distributed Computing - ISPDC 2012, p. 103–110. IEEE, Munich, June 2012.
• W. Eckhardt and A. Heinecke: An efficient Vectorization of Linked-Cell Particle Simulations [BibTeX].
  In ACM International Conference on Computing Frontiers, p. 241–243, May 2012.
• C. Niethammer, C. W. Glass, M. Bernreuther, S. Becker, T. Windmann, M. T. Horsch, J. Vrabec and W. Eckhardt: Innovative HPC Methods and Application to Highly Scalable Molecular Simulation (IMEMO) [BibTeX].
  In Inside - Innovatives Supercomputing in Deutschland, Volume 10(1), April 2012.

Volume 2010

• H.-J. Bungartz, W. Eckhardt, T. Weinzierl and C. Zenger: A Precompiler to Reduce the Memory Footprint of Multiscale PDE Solvers in C++ [] [BibTeX].
  In P.M.A. Sloot (ed.), Future Generation Computer Systems, Volume 26(1), p. 175–182. Elsevier, January 2010.
• W. Eckhardt and T. Weinzierl: A Blocking Strategy on Multicore Architectures for Dynamically Adaptive PDE Solvers [pdf] [BibTeX].
  In R. Wyrzykowski, J. Dongarra, K. Karczewski and J. Wasniewski (ed.), Parallel Processing and Applied Mathematics, PPAM 2009, Volume 6068(1) of Lecture Notes in Computer Science, p. 567––575. Springer-Verlag, 2010.

Volume 2009

• W. Eckhardt: Automatisierte Rekursionseliminierung für einen dynamisch adaptiven PDE Löser [pdf] [BibTeX].
  Diplomarbeit, Institut für Informatik, Technische Universität München, May 2009.

Volume 2008

• H.-J. Bungartz, M. Mehl, T. Weinzierl and W. Eckhardt: DaStGen - A Data Structure Generator for Parallel C++ HPC Software [] [BibTeX].
  In Bubak, van Albada, Sloot and Dongarra (ed.), ICCS 2008: Advancing Science through Computation, Part III, Volume 5103 of Lecture Notes in Computer Science, p. 213–222. Springer-Verlag, Heidelberg, Berlin, June 2008.

Volume 2007

• W. Eckhardt: Design and Implementation of a Record Class Generator for HPC Applications [BibTeX].
  Studienarbeit/SEP/IDP, Fakultät für Informatik, Technische Universität München, November 2007.

Talks

Volume 2015

• N. Tchipev, A. Wafai, C. Glass, W. Eckhardt, A. Heinecke, H.-J. Bungartz and P. Neumann: Optimized Force Calculation of Molecular Dynamics Simulations for the Intel Xeon Phi [BibTeX].
  Euro-Par 2015: Parallel Processing Workshops, Vienna, August 2015.
• N. Tchipev, W. Eckhardt, A. Heinecke, P. Neumann and H.-J. Bungartz: SIMD Vectorization for Molecular Dynamics [BibTeX].
  ISC '15 BoF Exploiting SIMD-Parallelism - Today and Tomorrow, Frankfurt, July 2015.
• N. Tchipev, W. Eckhardt, P. Neumann and H.-J. Bungartz: Extending the Vectorization of ls1-mardyn [BibTeX].
  International Workshop Molecular Modelling and Simulation: Science, Engineering and Industrial Applications, Frankfurt, March 2015.
• M. Horsch, S. Werth, C. Niethammer, C. Glass, W. Eckhardt, P. Neumann, H.-J. Bungartz, J. Vrabec, H. Hasse and T. Neckel: Computational Molecular Engineering: An Emerging Technology in Process Engineering [BibTeX].
  SIAM CSE, Salt Lake City, March 2015.

Volume 2014

• W. Eckhardt, N. Tchipev and P. Neumann: ls1 mardyn: Vectorisation, Multicore Exploitation and Large-Scale Distributed-Memory Parallelisation [BibTeX].
  Workshop Exploiting Heterogeneous Multi-Core and Many-Core Platforms for Atomic and Molecular Simulations, Daresbury Laboratory, September 2014.
• W. Eckhardt: Longrange Electrostatics: Ewald vs. ExaFMM [BibTeX].
  VI. Jahrestagung der Boltzmann-Zuse-Gesellschaft, Paderborn, June 2014.
• W. Eckhardt: Optimising a Molecular Dynamics Code for Xeon Phi: a Case Study [BibTeX].
  HP-Cast 22, Leipzig, June 2014.

Volume 2013

• W. Eckhardt: 591 TFLOPS Multi-trillion Particles Simulation on SuperMUC [pdf] [BibTeX].
  International Supercomputing Conference 2013, Leipzig, June 2013.
• W. Eckhardt: 591 TFLOPS Multi-trillion Particles Simulation on SuperMUC [BibTeX].
  PRACE Scientific Conference 2013, Leipzig, June 2013.
• W. Eckhardt: 591 TFLOPS Multi-trillion Particles Simulation on SuperMUC [BibTeX].
  V. Jahrestagung der Boltzmann-Zuse-Gesellschaft, Kaiserslautern, TU Kaiserslautern, June 2013.
• W. Eckhardt: The Simulation Code Mardyn - Recent Developments [BibTeX].
  International Workshop Molecular Modeling an Simulation: Natural Science meets Engineering, Frankfurt, March 2013.

Volume 2012

• W. Eckhardt: Efficient Implementation of MD Simulation [BibTeX].
  IV. Jahrestagung der Boltzmann-Zuse-Gesellschaft, Universität Paderborn, June 2012.
• W. Eckhardt: Memory-Efficient Implementation of a Rigid-Body Molecular Dynamics Simulation [pdf] [BibTeX].
  The 11th International Symposium on Parallel and Distributed Computing - ISPDC 2012, Garching, June 2012.
• W. Eckhardt: An efficient Vectorization of Linked-Cell Particle Simulations [BibTeX].
  ACM International Conference on Computing Frontiers. Ph.D. Forum., Cagliari, May 2012.
• W. Eckhardt: Memory-Efficient Implementation of a Rigid-Body Multi-Center Molecular Dynamics Simualtion for Nano-Fluidics [pdf] [BibTeX].
  ASIM TCSE, Wien, February 2012. Poster Presentation.

Volume 2011

• P. Neumann, P. Hoffmann, W. Eckhardt and J. Harting: All Good Things Come in Threes: Current Software and Algorithmic Developments for Molecular Dynamics-Lattice Boltzmann Simulations [pdf] [BibTeX].
  Cecam Workshop Multiscale Modeling of Simple and Complex Liquid Flow Using Particle-Continuum Hybrids, Zaragoza, October 2011. Poster Presentation.

Volume 2009

• W. Eckhardt and T. Weinzierl: A Blocking Strategy on Multicore Architectures for Dynamically Adaptive PDE Solvers [BibTeX].
  PPAM 2009, Wroclaw, Poland, September 2009.