Ghislain Lartigue

Design of implicit high-order filters on unstructured grids for the identification of large-scale features in large-eddy simulation and application to a swirl burner

The analysis of large-scale structures from highly refined unsteady simulations becomes challenging as the mesh resolution increases, and some new tools must be developed in order to perform their identification and extraction. A solution is to use filters to remove the smallest flow motions. High-order filters, characterized by their good selectivity properties, were implemented in an unstructured finite-volume solver for large-eddy simulation, and their ability to extract structures of a given scale was tested on canonical flows. Then, these filters were applied on an aeronautical swirl burner with a complex geometry. The results show that novel high-order filters are able to extract the precessing vortex core from this realistic turbulent flow. High-order filtering enables to study in detail this large-scale structure and to gain insight into the dynamic of swirl flows.

Using Large Eddy Simulations to Understand Combustion Instabilities in Gas Turbines

This paper presents a study of the stability of a swirled premixed combustion chamber both with and without reaction using Large Eddy Simulation and a numerical solver able to handle complex geometries. It is shown that cold flow instabilities, which are observed in the non-reacting case (especially precessing vortex cores), are strongly damped in the reacting case, suggesting that these natural flow instabilities are not the first source of combustion instabilities in such configurations.

CQA: A code quality analyzer tool at binary level

Most of todays performance analysis tools are focused on issues occurring at multi-core and communication level. However there are several reasons why an application may not correctly behave in terms of performance at the core level. For a significant part, loops in industrial applications are limited by the quality of the code generated by the compiler and do not always fully benefit from the available computing power of recent processors. For instance, when the compiler is not able to vectorize loops, up to a 8x factor can be lost. It is essential to first validate the core level performance before focusing on higher level issues. This paper presents the CQA tool, a loop-centric code quality analyzer based on a simplified unicore architecture performance modeling and on quality metrics. The tool analyzes the quality of the code generated by the compiler. It provides high level metrics along with human understandable reports that relates to source code. Our performance model assumes that all data are resident in the first level cache. It provides architectural bottlenecks and an estimation of the number of cycles spent in each iteration of a given innermost loop. Our modeling and analyses are statically done and requires no execution or recompilation of the application. We show practical examples of situations where our tool is able to provide very valuable information leading to a performance gain.

A multi-grid framework for the extraction of large-scale vortices in Large-Eddy Simulation

Abstract The analysis of large-scale vortices from highly refined unsteady simulations becomes challenging as the mesh resolution increases. Beyond the large amount of data that needs to be processed, classical vortex visualization techniques based on invariants of the velocity gradient tensor fail in extracting the large-scale vortices as the velocity gradient tensor magnitude is greater for small turbulent eddies than for energy-containing vortices. This problem is even more important in highly-resolved simulations with a broad range of eddies. The methodology presented here is a geometric multi-grid high-order filtering (MGHOF) framework for on-line analysis of high-fidelity simulations. This approach relies on high-order implicit filters and enables the extraction of large-scale features from Large-Eddy Simulations (LES) on massive and distributed unstructured grids at a reduced cost. The MGHOF framework is first described and validated, then the methodology is applied to a 3D turbulent plane jet and to the LES of a 3D low-Mach number turbine blade with various mesh sizes, ranging from a few million to a few billion tetrahedra. In the latter case, the MGHOF enables to perform the dynamic mode decomposition of the velocity and temperature fields for the finer grid resolution.

Modeling and Analysis of the Interactions of Coherent Structures with a Spray Flame in a Swirl Burner

With the constant increase in super-computing power, Large-Eddy Simulation (LES) has become an important tool for the modeling and the understanding of flame dynamics in complex burners. A fine description of the reaction layers in such devices requires fine meshes and the resolution of a broad range of turbulent scales. Unfortunately, extracting the large-scale features is not trivial. To this aim, implicit high-order filters that are based on simple low-order finite-volume operators have been proposed. These filters are applied in the LES of the MERCATO burner in order to study the complex interactions of the Precessing-Vortex Core, a large vortex typical of swirl burners, and a spray flame. High-order filters conveniently enable the analysis of the flame anchoring and its dynamics in the wake of the PVC.

Pre-exascale Architectures: OpenPOWER Performance and Usability Assessment for French Scientific Community

Exascale implies a major pre-requisite in terms of energy efficiency, as an improvement of an order of magnitude must be reached in order to stay within an acceptable envelope of 20 MW. To address this objective and to continue to sustain performance, HPC architectures have to become denser, embedding many-core processors (to several hundreds of computing cores) and/or become heterogeneous, that is, using graphic processors or FPGAs. These energy-saving constraints will also affect the underlying hardware architectures (e.g., memory and storage hierarchies, networks) as well as system software (runtime, resource managers, file systems, etc.) and programming models. While some of these architectures, such as hybrid machines, have existed for a number of years and occupy noticeable ranks in the TOP 500 list, they are still limited to a small number of scientific domains and, moreover, require significant porting effort. However, recent developments of new paradigms (especially around OpenMP and OpenACC) make these architectures much more accessible to programmers. In order to make the most of these breakthrough upcoming technologies, GENCI and its partners have set up a technology watch group and lead collaborations with vendors, relying on HPC experts and early adopted HPC solutions. The two main objectives are providing guidance and prepare the scientific communities to challenges of exascale architectures.

Design of High-Order Implicit Filters on Unstructured Grids for the Identification of Large-Scale Features in Large-Eddy Simulations

Large-Eddy Simulation (LES) and Direct Numerical Simulation (DNS) are increasingly popular modeling tools for the understanding and the prediction of turbulent flows.