High Performance Computing and Scientific Computing @ Durham's Computer Science
We study the Science of Computing behind Computational Sciences.
Scientific computing has become the foundation of many areas of science. It is computer simulations that allow various application areas to progress. In this context, high performance computing is a catalysator for insights-through-computing. With new hardware generations arriving, it emancipates from a nice-to-have feature into a mandatory craft in many disciplines.
Our research goes beyond application-specific number crunching, code analysis and tuning, i.e. it goes beyond the scientific computing and HPC research found in many application areas. Instead, it searches for innovative algorithms, algorithmic paradigms, patterns, and methodolgies to meet next generation’s supercomputing challenges, and it investigates into the foundations of scientific computing as a whole.
At the moment, our major contributions are made in the area of
- domain-specific languages for numerical simulations,
- algorithms behind efficient multiscale methods, and
- adaptive mesh refinement.
- Researchers from Durham’s CS HPC team are a driving force behind the development of Firedrake.
- Durham’s HPC team is the virtual home of the Peano framework.
- The European Union’s H2020 project ExaHyPE is actively driven by our HPC researchers.
The group actively collaborates with various compute-heavy departments in Durham. Highlights are joint projects, activities and research with the Department of Mathematics, the Physics department with its Institute for Particle Physics Phenomenology and its Institute for Computational Cosmology. Further links with Earth Sciences and the Department of Engineering yield interesting research. HPC’s most active interdisciplinary activities are typically organised under the umbrella of the Institute for Data Science.
Our researchers rely on Durham’s own supercomputer Hamilton, they have access to a 3x15m 3d visualisation wall, and we host several experimental workstations. By the end of 2019, we plan to purchase our own Mellanox BlueField cluster in collaboration with the DiRAC consortium.
Dr Tobias WeinzierlAssociate Professor
Mr Charles MurrayPhD student
Mr Meshal AlharbiPhD student
Mr Henry WestmacottPhD student
Dr Lawrence MitchellAssistant Professor
Peter NoblePhD Student
Dr Tobias Weinzierl
Associate Professor in the Department of Computer Science
T. Weinzierl: A Framework for Parallel PDE Solvers on Multiscale Adaptive Cartesian Grids. Verlag Dr. Hut, München, 2009.
M. Bader, H.-J. Bungartz and T. Weinzierl (ed.): Advanced Computing, Volume 93 of Lecture Notes in Computational Science and Engineering. Springer-Verlag, Heidelberg, Berlin, 2013.
Indicators of Esteem
- Programm Committee membership:
- ISC High Performance (2018)
- PASC (2016,2017)
- PARCO (2013)
- ISPDC (2012-2018)
- Visiting Researcher: Technische Universitaet Muenchen (TUM; 2014-2020) and KU Leuven (2012)
- Workshop organiser: ISC HPC 2016: Form Follows Function – Do algorithms and applications challenge or drag behind the hardware evolution?:See summary (https://arxiv.org/abs/1607.02835) which has been discussedf by several international newspages such as HPC Wire.
- Workshop organiser: ISC HPC 2018: The power of Lo(o)sing Control – When does a re-implementation of mature simulation fragments with HPC DSLs pay off?:
- Innovative Computing
- High-performance Computing
- Parallel Algorithms
- Scientific Computing
- Krestenitis, K. & Weinzierl, T. (2018). A Multi-Core Ready Discrete Element Method With Triangles Using Dynamically Adaptive Multiscale Grids. Concurrency and Computation: Practice and Experience
- Weinzierl, Marion & Weinzierl, Tobias (2018). Quasi-matrix-free hybrid multigrid on dynamically adaptive Cartesian grids. ACM Transactions on Mathematical Software 44(3): 32, 32:1-32:44.
- Reps, Bram & Weinzierl, Tobias (2017). Complex additive geometric multilevel solvers for Helmholtz equations on spacetrees. ACM Transactions on Mathematical Software 44(1): 2.
- Weinzierl, T., Verleye, B., Henri, P. & Roose, D. (2016). Two Particle-in-Grid Realisations on Spacetrees. Parallel Computing 52: 42-64.
- Weinzierl, Tobias, Bader, Michael, Unterweger, Kristof & Wittmann, Roland (2014). Block Fusion on Dynamically Adaptive Spacetree Grids for Shallow Water Waves. Parallel Processing Letters 24(3): 1441006.
- Weinzierl, Tobias & Köppl, Tobias (2012). A Geometric Space-time Multigrid Algorithm for the Heat Equation. Numerical Mathematics: Theory, Methods and Applications 5(1): 110-130.
- Weinzierl, Tobias & Mehl, Miriam (2011). Peano – A Traversal and Storage Scheme for Octree-Like Adaptive Cartesian Multiscale Grids. SIAM Journal on Scientific Computing 33(5): 2732-2760.
- Huckle, Thomas, Kallischko, Alexander, Roy, Andreas, Sedlacek, Matous & Weinzierl, Tobias (2010). An Efficient Parallel Implementation of the MSPAI Preconditioner. Parallel Computing 36(5-6): 273-284.
- Weinzierl, Tobias (2016). Multiscale Storage, Parallelisation and Programming Paradigms for Spacetrees in Scientific Computing. München: Technischen Universität München.
- Bader, Michael, Bungartz, Hans-Joachim & Weinzierl, Tobias (2013). Advanced Computing. Heidelberg, Berlin: Springer-Verlag.
- Weinzierl, Tobias (2009). A Framework for Parallel PDE Solvers on Multiscale Adaptive Cartesian Grids. München: Verlag Dr. Hut.
Chapter in book
- Bader, Michael & Weinzierl, Tobias (2015). Cache-Oblivious Spacetree Traversals. In Encyclopedia of Algorithms. Kao, Ming-Yang Springer. 1-6.
- Atanasov, Atanas, Srinivasan, Madhusudhanan & Weinzierl, Tobias (2012), Query-driven Parallel Exploration of Large Datasets, Large Data Analysis and Visualization (LDAV), 2012 IEEE Symposium on. 23 -30.
- Weinzierl, Tobias (2016). Form Follows Function – Do algorithms and applications challenge or drag behind the hardware evolution?. ISC High Performance 2016
- DaStGen – a C++ Data Structure Generator for HPC Software
- HPC Research at Durham’s School of Engineering and Computing Sciences
- Peano – a PDE solver framework
Mr Charles Murray
Mr Meshal Alharbi
(email at firstname.lastname@example.org)
- Coates, Graham, Li, Chunhui, Ahilan, Sangaralingam, Wright, Nigel & Alharbi, Meshal (2019). Agent-based modelling and simulation to assess flood preparedness and recovery of manufacturing small and medium-sized enterprises. Engineering Applications of Artificial Intelligence 78: 195-217.
Is supervised by
Mr Henry Westmacott
Dr Lawrence Mitchell
(email at email@example.com)
I am an Assistant Professor in the Department of Computer Science at Durham University. My research is in high performance computing and computational mathematics. Much of my recent focus has been in the development of compilers and software abstractions for the development of numerical models implemented using the finite element method. This research is concretely realised in the open source Firedrake project. I am particularly interested in preconditioning techniques for challenging problems in computational and atmospheric fluid dynamics.
The focus of my work is how to address the increasingly sophisticated needs of computational science practitioners by changing the way we think about numerical models. I develop computational mathematical abstractions that enable the automation of efficient implementations of complex, multiscale, numerical methods on modern supercomputers.
Compilers for numerical software
In the Firedrake project, I work on capturing the mathematical abstractions in numerical models, blending symbolic reasoning and numerical computation. This enables an approach to numerical software development that leverages symbolic information to synthesise high performance, parallel implementations of mathematical algorithms. This is possible through careful design of software abstractions, and development of domain-specific optimising compilers.
Fast solvers for geophysical flows
A large part of sophisticated numerical model development is in the design of robust linear and nonlinear solvers for the equations of interest. I have a particular interest in fast solvers for structure-preserving discretisations in atmospheric fluid dynamics. With Eike Müller, I developed a mesh-, and parameter- independent multigrid scheme for the mixed finite element discretisation proposed for the UK “GungHo” Dynamical Core project. We are presently working on multilevel schemes for the hybridised formulation of these equations, which should permit faster solvers. This latter work is in close collaboration with Colin Cotter, and Thomas Gibson.
- Innovative Computing
- Kirby, Robert C. & Mitchell, Lawrence (2018). Solver Composition Across the PDE/Linear Algebra Barrier. SIAM Journal on Scientific Computing 40(1): C76-C98.
- Kärnä, Tuomas, Kramer, Stephan C., Mitchell, Lawrence, Ham, David A., Piggott, Matthew D. & Baptista, António M. (2018). Thetis coastal ocean model: discontinuous Galerkin discretization for the three-dimensional hydrostatic equations. Geoscientific Model Development 11(11): 4359-4382.
- Homolya, Miklós, Mitchell, Lawrence, Luporini, Fabio & Ham, David A. (2018). TSFC: A Structure-Preserving Form Compiler. SIAM Journal on Scientific Computing 40(3): C401-C428.
- Rathgeber, Florian, Ham, David A., Mitchell, Lawrence, Lange, Michael, Luporini, Fabio, Mcrae, Andrew T. T., Bercea, Gheorghe-Teodor, Markall, Graham R. & Kelly, Paul H. J. (2017). Firedrake: automating the finite element method by composing abstractions. ACM Transactions on Mathematical Software 43(3): 24.
- Yamazaki, Hiroe, Shipton, Jemma, Cullen, Michael J.P., Mitchell, Lawrence & Cotter, Colin J. (2017). Vertical slice modelling of nonlinear Eady waves using a compatible finite element method. Journal of Computational Physics343: 130-149.
- Bercea, Gheorghe-Teodor, McRae, Andrew T. T., Ham, David A., Mitchell, Lawrence, Rathgeber, Florian, Nardi, Luigi, Luporini, Fabio & Kelly, Paul H. J. (2016). A structure-exploiting numbering algorithm for finite elements on extruded meshes, and its performance evaluation in Firedrake. Geoscientific Model Development 9(10): 3803-3815.
- McRae, A. T. T., Bercea, G.-T., Mitchell, L., Ham, D. A. & Cotter, C. J. (2016). Automated Generation and Symbolic Manipulation of Tensor Product Finite Elements. SIAM Journal on Scientific Computing 38(5): S25-S47.
- Lange, Michael, Mitchell, Lawrence, Knepley, Matthew G. & Gorman, Gerard J. (2016). Efficient Mesh Management in Firedrake Using PETSc DMPlex. SIAM Journal on Scientific Computing 38(5): S143-S155.
- Mitchell, Lawrence & Müller, Eike Hermann (2016). High level implementation of geometric multigrid solvers for finite element problems: Applications in atmospheric modelling. Journal of Computational Physics 327: 1-18.
- Guo, Xiaohu, Lange, Michael, Gorman, Gerard, Mitchell, Lawrence & Weiland, Michèle (2015). Developing a scalable hybrid MPI/OpenMP unstructured finite element model. Computers & Fluids 110: 227-234.
- Mitchell, Lawrence, Sloan, Terence M., Mewissen, Muriel, Ghazal, Peter, Forster, Thorsten, Piotrowski, Michal & Trew, Arthur (2014). Parallel classification and feature selection in microarray data using SPRINT. Concurrency and Computation: Practice and Experience 26(4): 854-865.
- Neic, A., Liebmann, M., Hoetzl, E., Mitchell, L., Vigmond, E. J., Haase, G. & Plank, G. (2012). Accelerating Cardiac Bidomain Simulations Using Graphics Processing Units. IEEE Transactions on Biomedical Engineering59(8): 2281-2290.
- Plank, Gernot, Smith, Nicolas, Mitchell, Lawrence & Niederer, Steven (2011). Simulating Human Cardiac Electrophysiology on Clinical Time-Scales. Frontiers in Physiology 2: 14.
- Mitchell, L. & Ackland, G. J. (2009). Boom and bust in continuous time evolving economic model. The European Physical Journal B 70(4): 567-573.
- Mitchell, Lawrence & Cates, Michael E. (2009). Hawkes process as a model of social interactions: a view on video dynamics. Journal of Physics A: Mathematical and Theoretical 43(4): 045101.
- Mitchell, L. & Ackland, G. J. (2007). Strategy bifurcation and spatial inhomogeneity in a simple model of competing sellers. Europhysics Letters (EPL) 79(4): 48003.
- Farrell, Patrick E., Mitchell, Lawrence & Wechsung, Florian (2018). An augmented Lagrangian preconditioner for the 3D stationary incompressible Navier-Stokes equations at high Reynolds number.
- Ham, David A., Mitchell, Lawrence, Paganini, Alberto & Wechsung, Florian (2018). Automated shape differentiation in the Unified Form Language.
- Kirby, Robert C & Mitchell, Lawrence (2018). Code generation for generally mapped finite elements.