Articles tagged with Supercomputing
Researchers use advanced simulation to model large mass ratio black hole merger, predicting characteristics of ultimate merged black hole and its speed. The simulation's success could help plan future gravitational wave detectors and answer mysteries about black holes.
The University of Texas at Austin has been selected by the Department of Energy to establish a Multidisciplinary Simulation Center, focusing on developing next-generation exascale predictive simulation capabilities. Researchers will combine computational simulations and physical experiments to improve predictive power.
Scientists have published a new calculation to test for tiny differences between matter and antimatter, building on the 1963 Nobel Prize-winning experiment that observed a slight difference in kaon decays. The new calculation provides more accurate predictions for kaon decays and offers a way to search for effects beyond the Standard M...
The University of Texas at Dallas has gained access to high-performance computing resources through the donation of systems like Stampede1, increasing its research capabilities. The introduction of Ganymede has enabled researchers to conduct complex projects like seismic imaging and data science, reducing computation time by up to 5 days.
Biological condensates, previously known as membrane-less organelles, have been found to play a crucial role in DNA repair and aging. Researchers used the Frontera supercomputer to study their behavior and recruitment of molecules.
The EuroCC project, coordinated by the Gauss Centre for Supercomputing, will build a sustainable European HPC ecosystem. The network will be supported by CASTIEL, which will promote interaction and expertise exchange across the entire EuroCC network.
A new AI system has been developed to detect cryptocurrency miners hijacking supercomputers for illicit mining. The system compares programs based on graphs, identifying malicious codes and preventing computing power theft.
Researchers at the University of Delaware used supercomputing resources to gain insights into the hepatitis B virus's genetic blueprint and how its protein shell assembles itself. They found that a mutation impairs this assembly process, revealing communication between different regions of the protein.
Researchers at Washington University in St. Louis developed a new algorithm called Parallel Residual Projection (PRP) to solve linear inverse problems by breaking them down into smaller tasks that can be solved in parallel on standard computers.
A team of researchers found that unequal neutron-star mergers can create an electromagnetic signal, which could be detected using gravitational-wave detectors like LIGO. The simulations revealed that the larger star tears apart its partner, creating a slower merger and allowing an 'electromagnetic bang' to escape.
USTC successfully completed a large-scale solid system simulation with tens of thousands of atoms using the Sunway TaihuLight supercomputer. The achievement demonstrates the power of domestic high-performance parallel computing software and hardware.
A team of scientists has confirmed that liquids show elasticity and rigidity like solids at the microscopic scale, contrary to long-held assumptions. The study's findings have significant implications for nanodevices, nanostructures, and metamaterials.
Fugaku achieved a LINPACK score of 415.53 petaflops, surpassing its nearest competitor Summit in the US, and ranked first on HPCG and Graph 500. The supercomputer will address high-priority social and scientific issues through various applications.
Researchers at KU Leuven have created a self-consistent simulation of solar flares, allowing them to calculate the energy conversion efficiency. This breakthrough enables the prediction of key aspects of space weather phenomena, including the Northern Lights.
A joint research team from Hong Kong, Beijing, and Shanghai discovered a rare-earth magnet TmMgGaO4 (TMGO) that realizes the long-sought-after two-dimensional topological Kosterlitz-Thouless (KT) phase. The breakthrough provides the missing piece of KT phenomena in bulk magnetic materials.
The COMPASS experiment at CERN is analyzing the proton's inner structure using particle collisions and complex algorithms. The team confirmed a theoretically expected sign change in the Sivers function, which relates to quark orbital motion inside the proton.
Sandia will be one of the first DOE laboratories to receive Fujitsu's new A64FX processor, optimized for memory-speed bottleneck breakage. The 48-core processor provides greater fractions of usable peak performance and supports collaboration with the Japanese supercomputing community.
Scientists developed a computational pipeline to screen drugs for induced arrhythmias, distinguishing between proarrhythmic agents and safe ones. The pipeline uses multi-scale computer simulation data to predict proarrhythmia vulnerability, enabling the identification of potentially toxic compounds.
Researchers are utilizing HPC to understand the virus at a molecular level, identify potential treatments, and accelerate vaccine development. Epidemiologists are also using supercomputers to model disease spread and predict hotspots, guiding policy makers' decisions in containing the pandemic.
Using digital twins, researchers simulate the complex interactions within urban environments to predict how changes in design could affect life there. The team developed a comprehensive model of Herrenberg city using space syntax, GIS data, and traffic control systems.
A study published in the Astrophysical Journal reveals that massive galaxies attain their size by merging with smaller ones. Researchers used a combination of observation and modelling to analyze how gases within galaxies move, finding evidence that many stars have been acquired from outside.
The new method makes it possible to create significantly more accurate fuel models for nuclear power plants. The researchers used atomistic models of the material comprising hundreds of thousands of atoms and supercomputers to calculate their trajectories over hundreds of millions or even billions of integration steps.
Scientists used molecular dynamics simulations to understand how SDS causes protein unfolding, revealing microscopic details of the process. The study provides insights into the properties of SDS-protein interactions and their applications in protein sequencing.
The Texas Advanced Computing Center has announced that the National Science Foundation has approved allocations of supercomputing time on Frontera to 49 science projects. These projects will utilize a total of 54 million node hours and constitute approximately 65% of the total time on the system being allocated for this year.
Researchers have created a massive computer model of the coronavirus, which will help design new drugs and vaccines. The model, built by Rommie Amaro's team, contains 200 million atoms and simulates the virus's interaction with human cells.
Researchers used supercomputer simulations to study viral reproduction and DNA replication mechanisms. They discovered that twisting stress in protein filaments plays a key role in creating membrane deformations, which is crucial for virus release and cellular processes.
Scientists used XSEDE-allocated supercomputers to study ion transport through nanoporous membranes. Advanced path sampling techniques captured the kinetics of solute transport, revealing a previously unknown mechanism called induced charge anisotropy that affects ion movement.
The Hawk supercomputer boasts a peak performance of 26 Petaflops, enabling cutting-edge academic and industrial research in areas like energy efficiency, climate modeling, and pandemic research. The system will also support the digitalization of industry in Baden-Württemberg and Germany.
A thermosyphon cooling unit installed at Sandia National Laboratories' supercomputer center saved 554,000 gallons of water during its first six months of operation. The unit also reduced electricity consumption by 195,000 kilowatt hours, making it a cost-effective solution for cooling large servers.
Researchers have developed a new computational tool for simulating fusion plasmas, which could help hold and squeeze superheated gas in extreme conditions. Additionally, scientists at Oak Ridge National Laboratory have created an approach to scan massive datasets of large-scale satellite images more efficiently, allowing for faster map...
Researchers developed a novel machine learning procedure that resolved Paleozoic biodiversity to within 26,000 years, clarifying diversification and extinction events. The improved resolution revealed new aspects of Paleozoic biodiversity and showed a correlation between atmospheric carbon dioxide levels and paleobiodiversity change.
Scientists have determined the structure of the B-Raf protein, which is responsible for about 50% of melanomas. The study reveals an asymmetric organization of the complex, enabling asymmetric activation of the B-Raf dimer, a mechanism that explains the origin of paradoxical activation by small molecule inhibitors.
The ETH Zurich team's project, 'A Data-Centric Approach to Extreme-Scale Ab initio Dissipative Quantum Transport Simulations,' developed a new framework called DaCe OMEN for simulating heat in transistors. The simulation achieved a two-orders-of-magnitude speedup and can help design better computer chips.
The DOE's Innovative and Novel Computational Impact on Theory and Experiment program has awarded supercomputer access to 47 science projects for 2020. Projects will explore topics such as galaxy formation, molecular dynamics, and engineering simulations.
A team from University of Tokyo utilized Summit's AI architecture to develop a faster solver for earthquake simulations, enabling more accurate models. The new approach accelerated simulation times by a factor of 1000, improving the efficiency and reliability of earthquake modeling.
Researchers at DOE/Oak Ridge National Laboratory develop novel parallel strategy for turbulent flow simulation on Summit. The approach successfully solves large-scale problems, revolutionizing the field of computational fluid dynamics.
Researchers at Oak Ridge National Laboratory have developed a building simulator to test energy savings in various buildings. The simulation is being tested in Chattanooga, Tennessee, as part of a partnership between the DOE and Electric Power Board.
Researchers used physics-informed generative adversarial networks (GANs) to model subsurface flow in the Hanford Site, achieving exaflop performance. The approach enabled estimation of hydraulic conductivity and hydraulic head with high accuracy, overcoming the limitations of traditional methods.
The Tapis Project aims to simplify access to powerful supercomputers and manage data from diverse sources. The platform offers event-driven computing, streamlining workflows and enabling hands-free analysis.
The researchers identified underlying causes of the deadly Palu earthquake and tsunami using coupled computer models. The team found that the earthquake-induced movement of the seafloor beneath Palu Bay itself could have generated the tsunami, meaning landslides contributed less to its formation than previously thought.
Frontera, the fastest academic supercomputer in the world, is set to revolutionize scientific research in the US. With its leadership-class computing capability, Frontera will support complex science applications in fields like black hole physics, climate modeling, and drug design.
Milinda Fernando and Staci Smith receive the fellowship for their work on high-performance algorithms and dynamic re-routing algorithms, respectively. Their research enables efficient use of modern supercomputers in various scientific disciplines.
Researchers at the University of Edinburgh used supercomputer calculations to study the growth and collapse of tiny bubbles, known as nanobubbles. The findings suggest that these bubbles can burst to release powerful jets of liquid, causing damage to industrial structures.
Frontera, located at the University of Texas at Austin, achieved the highest scale and data analysis capabilities ever deployed at a university in the US. The system supports dozens of research teams aiming to solve massive computational problems, including climate simulations and machine learning-enabled cancer studies.
The US Department of Energy is investing $32 million to accelerate material design through supercomputers. Seven projects will develop open-source software for designing functional materials with various applications.
Researchers used supercomputers to simulate complex earthquake ruptures, documenting interactions between faults and analyzing results with advanced visualization software. The model helps understand how faults interact during earthquake rupture, enabling scientists to study past earthquakes and possible future scenarios.
Researchers at Oak Ridge National Laboratory developed a stretchy plant-derived material with superior adhesion properties, while also advancing additive manufacturing techniques for metal parts. Additionally, they utilized the Summit supercomputer to simulate small modular nuclear reactor designs with improved efficiency.
A team of researchers at Virginia Tech has created a new system to efficiently distribute data processing tasks across thousands of servers in supercomputers, achieving balanced loads and improved performance. The novel technique uses machine learning to predict task types and amounts, allowing for optimized load balancing.
Researchers designed proteins that can assemble into complex structures using supercomputers and artificial charges. The stacked octamer structure consists of 16 proteins, resembling a braided ring with highly ordered and specific interactions.
Researchers from Rochester Institute of Technology and Georgia Tech developed a new approach to simulate the heart's electrophysiology in real-time, allowing doctors to better discuss life-threatening heart conditions with patients. This method uses WebGL code to repurpose graphics cards for faster scientific computing applications.
A team of geophysicists from LMU München used simulations to study the 2016 Kaikoura earthquake, which ruptured over 20 fault segments. The model showed that a weakly loaded fault was boosted by gradual slippage and low frictional resistance.
Researchers used supercomputer simulations to analyze interactions between tiny ripples on droplet surfaces, finding that these waves enable the initial contact and merger of droplets. The study's findings have implications for improving 3D printing technologies and understanding thunderstorm formation.
The Argonne National Laboratory's Aurora supercomputer will revolutionize scientific research and discovery with its exaFLOP performance and ability to handle both HPC and AI. The system is expected to have a significant impact on various fields, including cancer research, climate modeling, and veterans' health treatments.
Researchers at Linköping University have developed a theoretical model that simulates the degradation of hard cutting materials. The model, published in Materials journal, enables the manufacturing industry to save time and money by developing tools with greater hardness and resistance.
The NSF awards funding to further develop Open OnDemand, a platform providing web-based access to high-performance computing services. The project aims to simplify access to HPC resources, making it more accessible to new users and disciplines.
Researchers at Vanderbilt University Medical Center used sophisticated gene sequencing and computing techniques to analyze antibody-producing white blood cells. They found a surprisingly high frequency of shared clonotypes, which could aid in developing universal vaccines and treatments.
Scientists have sequenced a key part of the human immune system, discovering unexpected overlaps between adults and infants' antibody sequences. This finding could provide potential new targets for vaccines and treatments that work across populations.
By employing high-performance computing, researchers have developed new models for fine-scale turbulence data that can be used to inform large-eddy simulations, bringing accurate jet spray simulations to a commercial level. This advancement aims to improve fuel injection efficiency and spraying accuracy in various industrial processes.
By combining experimental results with simulations, researchers can gain insights into the atomic structure of 2D materials like graphene. This breakthrough could lead to the development of more efficient batteries and other electronics.
Researchers at Argonne National Laboratory are using machine learning algorithms to optimize engine simulations, significantly reducing design time and increasing accuracy. The project aims to create a more efficient and emissions-free combustion process, with potential applications in the automotive industry.