PRACEdays17 Posters

PRACEdays17 Best Poster Award

For the first time the PRACE Best Student Poster Award was presented. Martin Vögele (Max Planck Institute of Biophysics) received the prize for his poster entitled “ Finite size effects on the dynamics in simulations of lipid membranes“.

The PRACE Best Poster Award was awarded to Felipe Nathan Oliveira (Barcelona Supercomputing Center) for his poster entitled “Nonlinear electromagnetic stabilization of ITG microturbulence by ICRF-driven fast ions in ASDEX Upgrade”.

 

Nonlinear electromagnetic stabilization of ITG microturbulence by ICRF-driven fast ions in ASDEX Upgrade



Tokamaks are devices used to confine a fusion fuel mixture heated to high temperatures in the range of hundreds of millions of degrees, arriving in a state of matter called plasma. Several auxiliary systems are used for this purpose. They include Neutral Beam Injection (NBI) and electromagnetic waves. The largest tokamak up-to-date, aiming to achieve self-sustained nuclear fusion, is the International Thermonuclear Experimental Reactor (ITER) shown in figure 1. It is presently under construction in Cadarache, France.

DNS of Gravity-Driven Bubbly Flows



Present research is focused in combining High Performance Computing (HPC) and Direct Numerical Simulation (DNS) of gas-liquid multiphase flows, in order to study isothermal gravity-driven and thermocapillary-driven (g = 0) bubbly flows in a vertical channel. The Navier-Stokes equations and energy equation are discretized by means of the finite-volume approach, the pressure-velocity coupling is solved using a fractional-step projection method, whereas the conservative level-set (LS) methodology is used for interface capturing, thus, mass conservation issue that is known to affect standard LS methods, is circumvented. On the other hand, one of the disadvantages of interface capturing methods is the numerical and potentially unphysical coalescence of the fluid-interfaces. In this work, a novel multiple marker level-set methodology is used in order to avoid the numerical coalescence of the bubbles, indeed, long time simulations of bubbly flows including bubble collisions can be simulated.

POP: A Transversal HPC Centre of Excellence in Performance Optimisation and Productivity



The Performance Optimisation and Productivity Centre of Excellence in Computing Applications (POP) has received funding from the European Commission to uncover inefficiencies and their causes in HPC applications. From its start date in October 2015, POP aims to analyse 150 HPC codes in its first 2.5 years and provide  3M worth of savings through improved performance. This is achieved through free of charge performance investigations.

ECOSCALE Project: Energy-Efficient Heterogeneous COmputing at exaSCALE



Microscopic dynamics and topology of polymer rings immersed in a host matrix of longer linear polymers: Results from a detailed molecular dynamics simulation study and comparison with experimental data



High Performance Computing: an appealing solution for assessing radionuclide transport in fractured media at a grain-pore scale



Studying of statistical parameters for turbulent wake of a single wind turbine



Wind energy is animportant component of renewable energy sources. Studying of turbulent wakes for wind turbines and interactions with Atmospheric Boundary Layer is an urgent problem. Theory of chaos and fractal geometry with some research in fluid dynamics will change profoundly your vision of things in theory of turbulence. For this reason, we present some preliminary ideas and some results of the analysis of the fractal behaviour the evolution of statistical turbulence parameters in the wake of a single wind turbine through the multifractal analysis using the Imacal program.

For the first time the PRACE Best Student Poster Award was presented. Martin Vögele (Max Planck Institute of Biophysics) received the prize for his poster entitled “Finite size effects on the dynamics in simulations of lipid membranes”

Finite size effects on the dynamics in simulations of lipid membranes



Towards Embedded GPUs in HPC: A “green” counterpart of server GPUs or something more?



Main Memory Provisioning for High-Performance Computing



Development of a New Deterministic Neutron Transport Code for Fusion Applications



Share: Share on LinkedInTweet about this on TwitterShare on FacebookShare on Google+Email this to someone