A further increase in the performance of supercomputers is expected over the next few years. So-called exascale computers will be able to deliver more precise simulations. This leads to considerably more data. Fraunhofer SCAI develops efficient data analysis methods for this purpose, which provide the engineer with detailed insights into the complex technical contexts.
Simulations on supercomputers answer important industrial questions, such as how air flows behave in air conditioning systems, on rotor blades or for entire airplanes.
For less noise: high-resolution simulation of flows in a car air conditioning system
© Fraunhofer SCAI
In the coming years, new approaches to hardware design will make computers available which have a capacity of one trillion computing operations per second. So-called exascale computers have a computing speed a thousand times higher than previous supercomputers. This enables them to overcome complex problems that are not yet solvable.
In addition, they deliver more accurate results and thus more economical solutions for industrial tasks. However, exascale simulations also generate significantly larger amounts of data. This creates new challenges in data analysis.
Scientists at the Fraunhofer Institute for Algorithms and Scientific Computing SCAI have addressed this problem. They develop methods which perform data analysis during simulation. This approach minimizes memory requirements and computing time. In this way, even huge amounts of data can be analyzed efficiently.
The SCAI scientists concentrate on methods for the comparative analysis of simulation data. For example, the comparison of the data enables engineers to see at a glance how a modified component shape affects the flow behaviour inside an air conditioning system and thus the aerodynamic noise generation level. Eventually, that will help to design better products.
As a member of the newly founded "Competence Center of the European Union for Engineering Applications" EXCELLERAT, Fraunhofer SCAI works together with 13 partners from 7 European countries.
The aim is to make the best possible use of the new potentials of supercomputing and thus strengthen Europe's competitiveness in this field. Over the next three years, the participants will deal with six selected program codes and associated workflows, such as data analysis, management and visualization.
In addition to the efficient handling of large amounts of data, the project focuses on the scalability of the codes. In this context, scalability means that programs work efficiently on both laptops and exascale computers.
Since EXCELLERAT should also help to expand the use of supercomputing in industrial contexts, the project team will optimize the codes using twelve application examples from industry. Among others, these include the simulation of flows in car air conditioning systems as well as the calculation of combustion chamber emissions. Moreover, a common knowledge platform is to be developed.
Prof. Dr. Jochen Garcke
Head of Department »Numerical Data-Driven Prediction«
Phone +49 2241 14-2286
Dipl.-Journalist (TU Dortmund) Michael Krapp | Fraunhofer-Institut für Algorithmen und Wissenschaftliches Rechnen SCAI
Multifunctional e-glasses monitor health, protect eyes, control video game
28.05.2020 | American Chemical Society
Researchers incorporate computer vision and uncertainty into AI for robotic prosthetics
28.05.2020 | North Carolina State University
Humans rely dominantly on their eyesight. Losing vision means not being able to read, recognize faces or find objects. Macular degeneration is one of the major...
In meningococci, the RNA-binding protein ProQ plays a major role. Together with RNA molecules, it regulates processes that are important for pathogenic properties of the bacteria.
Meningococci are bacteria that can cause life-threatening meningitis and sepsis. These pathogens use a small protein with a large impact: The RNA-binding...
An analysis of more than 200,000 spiral galaxies has revealed unexpected links between spin directions of galaxies, and the structure formed by these links...
Two prominent X-ray emission lines of highly charged iron have puzzled astrophysicists for decades: their measured and calculated brightness ratios always disagree. This hinders good determinations of plasma temperatures and densities. New, careful high-precision measurements, together with top-level calculations now exclude all hitherto proposed explanations for this discrepancy, and thus deepen the problem.
Hot astrophysical plasmas fill the intergalactic space, and brightly shine in stellar coronae, active galactic nuclei, and supernova remnants. They contain...
In living cells, enzymes drive biochemical metabolic processes enabling reactions to take place efficiently. It is this very ability which allows them to be used as catalysts in biotechnology, for example to create chemical products such as pharmaceutics. Researchers now identified an enzyme that, when illuminated with blue light, becomes catalytically active and initiates a reaction that was previously unknown in enzymatics. The study was published in "Nature Communications".
Enzymes: they are the central drivers for biochemical metabolic processes in every living cell, enabling reactions to take place efficiently. It is this very...
19.05.2020 | Event News
07.04.2020 | Event News
06.04.2020 | Event News
05.06.2020 | Life Sciences
05.06.2020 | Physics and Astronomy
05.06.2020 | Life Sciences