High-performance computing (HPC) specialists are looking forward to the technological improvements that should arrive in the coming years as supercomputers approach the exascale. New approaches in hardware design (including new processors and high-bandwidth memory) and in application development (for example, code parallelization and data processing) will expand the power of supercomputing and therefore make it possible to solve new kinds of complex problems. Continue reading
Current practice in the design of hydro-power plants is to determine empirically the most suitable design in a series of time-consuming experiments. However, SMEs in this sector have to face private and public tenders to sell their turbines in both national and global markets, where the competition is very strong and the development time is very short. The challenge facing Zeco was therefore to improve their design processes to remain competitive. Continue reading
Thereâ€™s no getting around High Performance Computing (HPC) for RECOM Services, a Stuttgart-based small and medium-sized enterprise (SME) for computational process optimization and problem analysis in industrial combustion. Their specifically developed 3D-simulationssoftwareÂ RECOM-AIOLOS is able to illustrate combustion processes in virtual reality without disturbing the ongoing operation. Naturally, both engineering and HPC know-how is needed in order to succeed. Continue reading
An in-situ visualization of the cardiac model implemented with ALYA code and NVIDIA IndeX scalable software solution demonstrates the power of HPC simulations in GPU-accelerated clusters
The demonstration has been shown with a real-time high-quality rendering at NVIDIA GTC17 Conference
The French SME, PRINCIPIA, located in southern France, manages to add to its portfolio an innovative service that helps improve the control of the dynamical behaviour of offshore structures while strongly reducing computing time. The La Ciotat-based company, which specializes in applied engineering for the offshore sector (oil and gas and renewable marine energy) and is the main developer of reference software in this field, has successfully adapted one of its simulation codes, DEEPLINES, to run on a high performance computing (HPC) system. Continue reading
In a collaborative work with CERFACS, Institut de MÃ©canique des Fluides de Toulouse, Centralesupelec, ONERA, DLR and TU Munchen, researchers have been able to use the CERFACS LES code AVBPÂ to simulate the instabilities which appear in certain rocket engines under extreme conditions. The simulation was performed thanks to a PRACE allocation of 80 M hours. The massively parallel capacities offered by AVBP have allowed the code to run on 65 536 processors in production mode (the code itself scales up to 200 000 cores) on the Bluegene Q machine of CINECA. Continue reading
Scania â€“ a very large and well-known Swedish automotive company â€“ establishes a partnership with PDC Center for High Performance Computing at the KTH Royal Institute of Technology to further improve the efficiency and performance of its vehicles using state of the art simulation and testing techniques.
ANSYS, the High Performance Computing Center (HLRS) of the University of Stuttgart and Cray Inc. have set a new supercomputing world record enabling organizations to create complete virtual prototypes of products faster than ever. ANSYS Fluent has been scaled to 172,032 computer cores on the HLRS supercomputer Hazel Hen, a Cray XC40 system, running at 82 percent efficiency. This is nearly a 5x increase over the record set two years ago when Fluent was scaled to 36,000 cores.
Global equipment manufacturers in the chemical and oil and gas industry, such as Sulzer Chemtech, often rely on commercial Computational Fluid Dynamics (CFD) software tools for the design of their equipment. These commercial codes are currently unable to handle complex twophase flows which exhibit challenging interfaces between gas and liquids such as travelling waves. The formation of interfacial waves, their frequency and amplitude are particularly difficult to model in industrial environments.