The Energy-oriented Centre of Excellence (EoCoE-II)

The EoCoE-II project, funded by the European Commission, applies cutting-edge computational methods in its mission to accelerate the transition to the production, storage and management of clean, decarbonized energy. This world-class consortium of 18 complementary partners from 7 countries, amounting to a unique network of expertise in energy science, scientific computing and HPC, intends to play its part in the reinforcement of a pan-European high performance computing infrastructure, benefitting researchers, key commercial players and SMEs alike.

Newsletter no.3

Dear subscribers,
we hope that this new issue of our newsletter finds you well. EoCoE-II activities are continuing despite the pandemia, and we are trying to keep all our commitments and our schedule unchanged. Below a selection of significant results, and information on our webinar offer, which was intensified to allow better participation and involvement from home.


Success stories

  • Wind for energy
  • HPC performance
  • Fusion


  • Numerical challenges for the simulation of magnetic fusion plasmas
  • Accelerated hydrologic modeling: ParFlow GPU implementation
  • Paving the way to AMR groundwater simulations with Parflow+p4est

EoCoE is hiring

Success stories: Wind for energy

Wall-modeled large-eddy simulation in a finite element framework
Herbert Owen, Georgios Chrysokentis, Matias Avila, Daniel Mira, Guillaume Houzeaux, Ricard Borrell, Juan Carlos Cajas, Oriol Lehmkuhl
Abstract: This work studies the implementation of wall modeling for large-eddy simulation in a finite element context. It provides a detailed description of how the approach used by the finite volume and finite differences communities is adapted to the finite element context. The new implementation is as simple and easy to implement as the classical finite element one, but it provides vastly superior results. In the typical approach used in finite elements, the mesh does not extend all the way to the wall, and the wall stress is evaluated at the first grid point, based on the velocity at the same point. Instead, we adopt the approach commonly used in finite differences, where the mesh covers the whole domain and the wall stress is obtained at the wall grid point, with the velocity evaluated at the first grid point off the wall. The method is tested in a turbulent channel flow at Re! = 2003, a neutral atmospheric boundary layer flow, and a flow over a wall-mounted hump, with significant improvement in the results compared to the standard finite element approach. Additionally, we examine the effect of evaluating the input velocity further away from the wall, as well as applying temporal filtering on the wall-model input.

Int J Numer Meth Fluids. 2020;92:20–37. https://doi.org/10.1002/fld.4770

Success stories: HPC Performance

Performance drop at executing communicationintensive parallel algorithms
José A. Moríñigo, Pablo Garcia‑Muller, Antonio J. Rubio‑Montero, Antonio Gomez‑Iglesias, Norbert Meyer, Rafael Mayo‑Garcia
Rafael Mayo-Garcia
Abstract: This work summarizes the results of a set of executions completed on three fat-tree network supercomputers: Stampede at TACC (USA), Helios at IFERC (Japan) and Eagle at PSNC (Poland). Three MPI-based, communication-intensive scientific applications compiled for CPUs have been executed under weak-scaling tests: the molecular dynamics solver LAMMPS; the finite element-based mini-kernel miniFE of NERSC (USA); and the three-dimensional fast Fourier transform mini-kernel bigFFT of LLNL (USA). The design of the experiments focuses on the sensitivity of the applications to rather different patterns of task location, to assess the impact on the cluster performance. The accomplished weak-scaling tests stress the effect of the MPI-based application mappings (concentrated vs. distributed patterns of MPI tasks over the nodes) on the cluster. Results reveal that highly distributed task patterns may imply a much larger execution time in scale, when several hundreds or thousands of MPI tasks are involved in the experiments. Such a characterization serves users to carry out further, more efficient executions. Also researchers may use these experiments to improve their scalability simulators. In addition, these results are useful from the clusters administration standpoint since tasks mapping has an
impact on the cluster throughput.

The Journal of Supercomputing 2020. https://doi.org/10.1007/s11227-019-03142-8

Success stories: Fusion

Linear collisionless dynamics of the GAM with kinetic electrons: comparison simulations/theory
V. Grandgirard, X. Garbet, Ch. Ehrlacher, A. Biancalani, A. Bottino, I. Novikau, Y. Asahi, E. Caschera, G. Dif-Pradalier, P. Donnel, Ph. Ghendrih, C. Gillot, G. Latu, Ch. Passeron, Y. Sarazin, D. Zarzoso
Abstract: Barely trapped and passing electrons have been recently predicted to strongly enhance the damping rate of Geodesic Acoustic Modes (GAMs) in tokamak plasmas, while keeping their real frequency almost unchanged as compared to the case with adiabatic electrons. In this paper, dedicated gyrokinetic simulations are successfully compared with these analytical predictions. Specifically, the scaling of the GAM damping rate with respect to the ion to electron mass ratio, to the electron to ion temperature ratio, to the safety factor, and to the aspect ratio is recovered in most regions of the relevant parameter space.

Phys. Plasmas 26, 122304 (2019); https://doi.org/10.1063/1.5113679

logo eocoe 120 scritta sotto 3 righe


Numerical challenges for the simulation of magnetic fusion plasmas
Date : 15/05/2020, 10.30AM
Speaker : Prof. Dr. Eric Sonnendrücker. Head of Numerical Methods in Plasma Physics Division at the Max Planck Institute for Plasma Physics

Abstract: Hosted by Prof. Eric Sonnendrücker of the Max-Planck Institute for Plasma Physics, this webinar will dive into magnetic fusion research, a scientific field pursued in order to provide a clean energy available on demand. The principle behind magnetic fusion research is to confine a plasma, which is a gas of charged particles at a very large temperature, around 100 000 degrees, so that the fusion reaction can generate energy with a positive balance. At such a high temperature, the plasma needs to be completely isolated from the wall of the reactor. This isolation can be achieved in toroidal devices thanks to a very large magnetic field. Due to the multiple and complex physical processes involved, theoretical research in this field relies heavily on numerical simulations and some problems require huge computational resources. After introducing the context of magnetic confinement fusion, we shall address different specific challenges for numerical simulations in this topic, which are in particular related to the multiple space and time scales that need to be spanned and to the geometry of the experimental devices. These can only be solved thanks to a close collaboration between physicists, mathematicians and HPC specialists. A few current research problems in this field going from the computation of a 3D equilibrium to fluid and kinetic simulations will be presented as an illustration.
Registration url: https://attendee.gotowebinar.com/register/7722234506387424779

Accelerated hydrologic modeling: ParFlow GPU implementation
Date : 10/06/2020, 10.30AM
Speaker : Jaro Hokkanen. Computer Scientist at Forschungszentrum Jülich

Abstract: Hosted by Jaro Hokkanen, computer scientist at Forschungszentrum Jülich, this webinar will address the GPU implementation of the Parflow code. ParFlow is known as a numerical model that simulates the hydrologic cycle from the bedrock to the top of the plant canopy. The original codebase provides an embedded Domain-Specific Language (eDSL) for generic numerical implementations with support for supercomputer environments (distributed memory parallelism), on top of which the hydrologic numerical core has been built. In ParFlow, the newly developed optional GPU acceleration is built directly into the eDSL headers such that, ideally, parallelizing all loops in a single source file requires only a new header file. This is possible because the eDSL API is used for looping, allocating memory, and accessing data structures. The decision to embed GPU acceleration directly into the eDSL layer resulted in a highly productive and minimally invasive implementation. This eDSL implementation is based on C host language and the support for GPU acceleration is based on CUDA C++. CUDA C++ has been under intense development during the past years, and features such as Unified Memory and host-device lambdas were extensively leveraged in the ParFlow implementation in order to maximize productivity. Efficient intra- and inter-node data transfer between GPUs rests on a CUDA-aware MPI library and application side GPU-based data packing routines. The current, moderately optimized ParFlow GPU version runs a representative model up to 20 times faster on a node with 2 Intel Skylake processors and 4 NVIDIA V100 GPUs compared to the original version of ParFlow, where the GPUs are not used. The eDSL approach and ParFlow GPU implementation may serve as a blueprint to tackle the challenges of heterogeneous HPC hardware architectures on the path to exascale.
Registration url: https://attendee.gotowebinar.com/register/9001459110700525835

Paving the way to AMR groundwater simulations with Parflow+p4est
Date : 01/07/2020, 11AM
Speaker : Jose Alberto Fonseca Castillo. Postdoctoral researcher at CEA/Maison de la Simulation

Abstract: The software library ParFlow is a complex parallel code that is used extensively for high performance computing, specifically for the simulation of surface and subsurface flow. The code discretizes the corresponding partial differential equations using cell centered finite differences on a uniform hexahedral mesh. Even with the current supercomputing resources, using uniform meshes may translate in prohibitively expensive computations for certain simulations. A solution to this problem is to employ adaptive mesh refinement (AMR) to enforce a higher mesh resolution only whenever it is required. To this this end, we have relegated ParFlow's mesh management to the parallel AMR library p4est. During this seminar, Jose Fonseca, postdoc researcher at CEA / Maison de la Simulation, will present the algorithmic approach used to perform this coupling and our latest efforts to generalize ParFlow's native discretization to the locally refined meshes obtained with p4est.
Registration url: https://attendee.gotowebinar.com/register/4162667866044575243

EoCoE is hiring

Engineer: High Performance Computing & Data Analysis
Requirement: Master degree or Engineer Degree Location: Grenoble Hosting Team: DataMove (Inria Grenoble)
Contact: Bruno Raffin (Bruno.Raffin@inria.fr)
Period: to start somewhere in 2020 (24 months)
More infos: https://lnkd.in/gPRYQZ8

The candidate will join the DataMove team located in the IMAG building on the campus of Saint Martin d’Heres (Univ. Grenoble Alpes) near Grenoble. The DataMove team is a friendly and stimulating environment gathering Professors, Researchers, PhD and Master students all leading research on High Performance Computing. Grenoble is a friendly city surrounded by the Alps mountains, offering a high quality of life and where you can experience all kinds of mountain related outdoors activities and more. This work is part of the EoCoE European project. As such, we will work in close collaborations with partners of the project, in particular with the Juelich Supercomputing Center, Germany, that develops the two applications targeted in this work.
Inria Grenoble
Read more on our website.
linkedin twitter youtube