SPP 1648: Software for Exascale Computing - Subproject: EXA-DUNE - Flexible PDE Solvers, Numerical Methods, and Applications (EXA-DUNE)

Basic data for this project

Type of projectSubproject in DFG-joint project hosted outside University of Münster
Duration at the University of Münster15/10/2012 - 31/03/2016 | 1st Funding period

Description

The aim of this interdisciplinary project, bringing together experts from the open source projects DUNE and FEAST, is to develop, analyse and realise new numerical, algorithmic and computational techniques to enable exascale computing for partial differential equations (PDEs) on heterogeneous massively parallel architectures. As the life time of PDE software is typically much longer than for hardware, flexible but nevertheless hardware-specific software components are developed based on the DUNE platform, which uses state-of-the-art programming techniques to achieve great flexibility and high efficiency to the advantage of a steadily growing user-community. Hardware-oriented numerical techniques of the FEAST project are integrated to optimally exploit the performance of the local (heterogeneous) nodes (multi-core multi-purpose CPUs, special purpose acceleration units like GPUs, etc.), w.r.t. specific structures of the given PDEs. The introduction of a hardware abstraction layer will make it possible to perform the necessary hardware-specific changes of essential components at compile time with at most minimal changes of the application code. Further adding to the great benefits from a combination of the strengths of DUNE and FEAST, modern numerical discretisations and solver approaches like adaptive multi-grid, localised spectral methods (e.g. higher-order Discontinous Galerkin schemes) and a hybrid parallel grid will increase the scalability. The EXA-DUNE toolbox is extended from petascale towards exascale level computing by introducing multi-level Monte Carlo methods for uncertainty quantification and multi-scale techniques which both add an additional layer of coarse grained parallelism, as they require the solution of many weakly coupled problems. The new methodologies and software concepts are applied to flow and transport processes in porous media (fuel cells, CO2 sequestration, large scale water transport), which are grand challenge problems of high relevance to society.

KeywordsPDE Solvers; Numerical Methods; Exascale Computing
Website of the projecthttp://www.sppexa.de/
Funding identifierEN 1042/2-1; OH 98/5-1
Funder / funding scheme
  • DFG - Priority Programme (SPP)

Project management at the University of Münster

Engwer, Christian
Ohlberger, Mario

Applicants from the University of Münster

Engwer, Christian
Ohlberger, Mario

Research associates from the University of Münster

Fahlke, Jorrit
Fritze, René
Kaulmann, Sven

Project partners outside the University of Münster

  • TU Dortmund University (TU Dortmund)Germany
  • Fraunhofer Institute for Industrial Mathematics (ITWM)Germany
  • Heidelberg UniversityGermany

Coordinating organisations outside the University of Münster

  • Technical University of Munich (TUM)Germany

Projects of the following funding period

Duration: 12/10/2015 - 31/12/2018 | 2nd Funding period
Funded by: DFG - Priority Programme
Type of project: Subproject in DFG-joint project hosted outside University of Münster

Related main project

Duration: 15/10/2012 - 31/12/2018
Funded by: DFG - Priority Programme
Type of project: Main DFG-project hosted outside University of Münster

Publications of the University of Münster resulting from the project

Mohring Jan, Milk Rene, Ngo Adrian, Klein Ole, Iliev Oleg, Ohlberger Mario, Bastian Peter (2015)
In: Lirkov I., Margenov S. D. , Wasniewski J. (eds.), Proceedings of 10th International Conference on Large-Scale Scientific Computations, Sozopol 2015145-152Springer International Publishing. doi:10.1007/978-3-319-26520-9_15
Research article in edited proceedings (conference) | Peer reviewed | Published
Kaulmann S, Flemisch B, Haasdonk B, Lie K, Ohlberger M (2015)
In: International Journal for Numerical Methods in Engineering5(102)1018-1040. doi:10.1002/nme.4773
Research article (journal) | Peer reviewed | Published

Doctorates resulting from the project

Efficient Schemes for Parameterized Multiscale Problems
Candidate: Kaulmann, Sven | Supervisors: Haasdonk, Bernard; Ohlberger, Mario
Period of time: until 16/12/2014
Doctoral examination procedure finished at: Doctoral examination procedure at another university (NOT University of Münster)