Thematic Workshop on HPC applications to turbulence and complex flows

Venue: University of Rome 'Tor Vergata'

Dates: 10 - 14 October 2016

Credit: 10 ECTS

Scientist-in-charge: Luca Biferale

 

Description:

The workshop on HPC applications to turbulence and complex flows is part of the series of thematic workshops offered by the HPC-LEAP project. The workshop focused on fundamental aspects of turbulence, micro- and nano-fluidic flows, kinetic modelling and Large-Eddy Simulations, including also lectures on modern SW/HW tools for High Performance Computing.

 
Topic Lecture hours
Fundamentals of Turbulent Flows 2
Dynamical systems approach 2
HPC applications to Eulerian and Langrangian Turbulence 2  
Fundamentals of LBM 2
Parallel Computing in fluids and statistical mechanics 2
Fundamentals of large-Eddy simulations 3
LBM for energy and environmental systems 2
Management tools for multi-developer codes 2
  17

 

Components:

  • Fundamentals of turbulent flows - Prof. Luca Biferale, UTOV: Fully developed turbulence (Eulerian) for homogeneous and isotropic flows, flows under rotations, strongly sheared flows. Intermittency and multifractals. Large deviation theory. Modern tools to analyze universal and non universal properties in turbulence using SO(3) decomposition.

  • Dynamical system theory - Dr Massimo Cencini, ISC-CNR: Dynamical system theory and applications to hydrodynamics. Tracers, inertial particles, active particles. Fractals, multifractal, Lyapunov exponents. Applications to study the preferential samples of particles in turbulent and laminar flows.

  • HPC applications to Eulerian and Lagrangian turbulence - Prof. Federico Toschi, TU/e: Pseudospectral codes, time integrators. Issues connected to massive simulations of a high number of particles. Boundary conditions for infinite-size particles and modelling of lubrication forces in Lattice Boltzmann methods.

  • Parallel computing in fluids and statistical mechanics - Dr Massimo Bernaschi, IAC-CNR: Basic concepts about GPU and Cuda language. Optimization of multi-GPU codes and shared memory. A set of examples about how to implement a CUDA program to integrate in parallel the diffusion equations.

  • Fundamentals of LBM - Prof. Mauro Sbragaglia, UTOV: LBM are mesoscopic methods, tracking the evolution in space and time of the probability distribution function for the particles, while the hydrodynamics of the Navier-Stokes (NS) equations is recovered from the coarse-grained behaviour of the system. Starting from the LBM, the key ingredients of the Chapman-Enskog expansion - allowing for the recovery of NS hydrodynamical equations were detailed. Recent advances of LBM, including the extensions to non-ideal fluids (both single-phase and multi-component), also in the presence of boundaries, were discussed. In the second part of the lectures, applications of LBM in the field of Soft-Matter were presented: (i) droplet break-up in presence of viscoelastic phases; (ii) droplet dynamics in presence of modulated wettability gradients; (iii) Soft-Glassy rheology in the presence of confinement. These applications showcased the successful extension of the traditional portfolio of LBM applications towards micro- and nanofluidics.

  • Tools to manage multi-developer codes - Dr F. Bonaccorso, UTOV: The lessons covered the basics and some advanced notions about the following topics: the Git versioning system; the CMake building system for source codes; the HDF5 portable file library for efficiently store the simulation data; the FFTW and P3DFFT libraries for high performance parallel FFTs.

  • LBM for the study of energy and environmental stystems - Dr G. Falcucci, UTOV: Details on multiphase flows in energy and environmental systems were discussed. The main aspects related to the development and testing of the 2–belt Equation of State (EOS) were presented. The preliminary results were reported and then the application to the 2–belt EOS to complex fluid-dynamic phenomena was presented. In particular, the implementation of a diesel engine fuel injector was discussed. The geometrical characteristics and the main fluid dynamic parameters affecting the internal fluid dynamics of the nozzle were presented and their implementation in 2D and 3D LBM schemes was discussed. The results of Direct Numerical Simulations of diesel fuel spray formation and break-up were displayed and discussed. The importance and ubiquity of fluid-structure interaction phenomena was presented. Then, starting from an experimental case provided by Harvard University, a complex, chemically-active porous medium was reconstructed within a Lattice Boltzmann framework. Preliminary simulations provided details on the location of the chemical conversion on the surface of the porous catalyst and highlight the inhomogeneity of the catalyst medium, in agreement with the experimental results. Further cases of fluid-structure interaction for phenomena of technical interest were proposed proving the versatility and reliability of LBM for this relevant arena.

  • Fundamentals of Large Eddy Simulations - Prof. Jim Brasseur, University of Colorado: Kolmogorov theory and scale separations. The sub-grid energy tensor. The issue of defining the filter. Effects of implicit and explicit filtering. Applications to homogeneous and isotropic turbulence, to wall bounded flows and to the Atmospheric boundary layer.


Links & Resources:

The agenda of the workshop can be downloaded here.

Further details and resources for the lectures given may be found on the website of the workshop

HPC-LEAP in a nutshell

Progress in computers and algorithms in the last years has made numerical simulation and modelling a key research methodology in both academia and industry, which in turn drives exascale computing in order to maintain excellence in research and innovation. A disruptive evolution in computer technologies is required for attaining exascale performances in the coming years bringing challenges that urgently need to be addressed across science and engineering fields. Therefore, new interdisciplinary strategies are required in order to educate the next generation of scientists to address such challenges enabling them to be at the forefront of their respective research fields. Instead of the traditional domain-specific training, integrated approaches are needed that can be best implemented by collaborative networks of universities, research institutes and industrial partners. HPC-LEAP is a highly interdisciplinary joint doctorate program realized by bringing together world-leading experts in applied mathematics, high performance computing technologies, particle and nuclear physics, fluid dynamics and life sciences to appropriately train researchers in Europe to exploit high performance computing, advance science and promote innovation. Students will be trained in mathematical and computational concepts underpinning current and future numerical simulations in turbulent flows, computational biology and lattice Quantum Chromodynamics. The research projects are designed to enhance collaborations and interactions across these disciplines, integrating non-academic partners, and to develop methodologies that efficiently use large-scale numerical simulations on future high performance computer systems. Students who complete this training program will be versatile to undertake highly interdisciplinary projects, well positioned to embark on a successful career in academia or the industrial sector.

Consortium Institutions

  • CyINew
         
         
  •            

  •      18      8
  • 9      10      11
  • 12    13          14 

  •