Articles tagged with Supercomputing
A new approach using deep learning speeds up supernova simulation by 99%, enabling more accurate modeling of galaxy evolution. This breakthrough could also apply to climate and earthquake simulations, providing valuable insights into complex phenomena.
Researchers have carried out the largest ever computer simulations to investigate the Universe's evolution, taking into account ordinary matter and dark energy. The FLAMINGO simulations provide a detailed picture of virtual galaxies and galaxy clusters, allowing for comparisons with observations from new high-powered telescopes.
Researchers from BSC and CSIC have developed an artificial protein capable of degrading PET micro- and nanoplastics with efficiency between 5 and 10 times higher than current PETases. The protein can be used as filters to purify or recycle plastics, offering a potential solution to environmental pollution.
The High Performance Data Facility Hub will provide researchers with unprecedented data management resources, accelerating scientific discovery through seamless access to large and complex datasets. The hub will be led by Jefferson Lab and partner with Lawrence Berkeley National Laboratory.
A Harvard team has successfully developed a self-correcting quantum computer using neutral atom arrays, achieving near-flawless performance with extremely low error rates. The breakthrough enables the creation of large-scale, error-corrected devices based on neutral atoms.
Researchers developed a scalable framework to quantify field-level agricultural carbon outcomes, integrating sensing techniques and advanced ecosystem models. The study provides a reliable method for measuring individual farm field-level greenhouse gas emissions and offers insights into the impact of farming practices on climate change.
Researchers aim to develop programmable formal specification-based data stream processor and hardware monitor to enhance microchip security and prevent malfunctions. They will explore novel technologies for real-time monitoring of physical and biological systems, including signal patterns within computer chips.
A recent study published in the Journal of Chemical Information and Modeling presents a significant breakthrough in accelerating giga-scale virtual screens using machine learning. The researchers successfully reduced processing time by 10-fold for 1.56 billion drug-like molecules, identifying top-scoring compounds in under ten days.
The Human Brain Project has achieved major breakthroughs in understanding the complex structure and function of the brain, enabling novel medical and technological applications. The project's digital atlas and personalized virtual models have provided unprecedented insights into brain conditions such as epilepsy and Parkinson's disease.
The collaboration has garnered $5.68 million of grant support, with a five-year $2.13 million National Institutes of Health R01 grant and a five-year $2.89 million NIH R01 grant. Researchers are using computational biology to model a key Alzheimer's protein called TREM2 and its interactions with small chemical molecules.
A Los Alamos-developed machine learning algorithm successfully processed massive data sets exceeding a computer's available memory. The algorithm divides data into manageable batches to prevent hardware bottlenecks, enabling efficient processing of large-scale applications in various fields.
The next generation of supercomputers is revolutionizing biophysics research, allowing researchers to simulate complex biological processes with extraordinary detail. This enables the creation of new proteins and design of novel molecular circuits, challenging longstanding biological assumptions.
Scientists discovered that climate shifts during the last 400,000 years influenced the frequency of Neanderthal-Denisovan interbreeding. The researchers found that temperature changes triggered habitat overlaps, leading to increased contact between the two species.
A recent study reveals that past climate changes and vegetation shifts played a key role in determining when and where early human species interbred. The research suggests that the overlap of habitats led to increased encounters and interactions among groups, increasing the chance of interbreeding.
Astronomers have observed and simulated the feeding of gas to three protostars in a trinary system, revealing how spiral arms, or 'streamers,' form and contribute to protostar growth. This study provides new insights into multi-star system formation, an important aspect of overall star formation theory.
A research group from Nagoya University simulated clear air turbulence using Japan's fastest supercomputer. They found that wind speed disturbances occur due to the collapse of Kelvin-Helmholtz instability waves, creating turbulence in the absence of visible clouds or other atmospheric disturbances.
Researchers created an isolated turbulent blob by firing vortex rings into a tank of water, allowing precise tracking of its parameters. This breakthrough enables scientists to study real-world turbulence more effectively, exploring questions about dissipation, expansion, and energy spread across scales.
Researchers developed a new technique called zero noise extrapolation (ZNE) that allows noisy quantum computers to produce accurate results for specific calculations. This breakthrough could enable the use of quantum computing for cutting-edge physics problems and improve classical algorithms.
ChatGPT consistently provided accurate, evidence-based answers in public health questions, beating rival AI assistants. Meanwhile, Amazon Alexa and Google Assistant struggled to recognize or refer users for help.
Researchers used a powerful supercomputer to analyze the 2019 Ridgecrest earthquakes, which showed how faults can interact across segments and cause unexpected interactions. The study's findings improve seismic hazard assessment and preparedness for complex earthquake systems.
The Oak Ridge Leadership Computing Facility has selected Matt Sieger as its new project director for the OLCF-6 effort. The system aims to provide leadership computing capabilities to researchers and is expected to be delivered in 2026.
Researchers at Oregon State University and Baylor University have made a breakthrough in reducing the energy consumption of photonic chips used in data centers and supercomputers. By using a new, ultra-energy-efficient method to compensate for temperature variations, they were able to reduce energy needs by over 1 million times.
Researchers are harnessing Frontier's exascale capabilities to tackle complex scientific challenges, including nuclear fusion and earthquake forecasting. The system has already demonstrated impressive performance, with updated rankings on the TOP500 list and improved benchmark scores.
Researchers simulated all known energy states of the carbon nucleus, providing insights into the puzzling Hoyle state and its configuration. The study reveals that protons and neutrons are clustered into groups, creating spatial formations with distinct shapes and energies.
A team of Chinese scientists developed high-resolution Earth system models incorporating clouds and ocean submesoscale eddies, capturing major weather-climate extremes. The models simulate cross-scale interactions and provide new insights into weather-climate mechanisms.
The Barcelona Supercomputing Center's new AI method combines CALIOPE-Urban predictions with urban data to provide accurate air quality maps and predict pollution limit exceedances. The study identifies the Eixample district as the worst-affected area in Barcelona, highlighting the need for effective policies to improve air quality.
Researchers developed a new framework for modeling task-switching, mimicking human behavior. The framework revealed two regions of the model's 'brain' doing each task, explaining the switch cost and potential benefits of splitting tasks.
Researchers used computer simulations to understand the formation of new subduction zones and the development of the Caribbean large igneous province. The study found that simultaneous subduction of two plates led to a major mantle flow, triggering the formation of a plume and extensive magmatic activity.
The Human Brain Project has achieved significant scientific breakthroughs, including world-leading 3D brain atlases and personalized medicine based on computational modeling. The project's legacy will leave a lasting impact on the research community.
A team of astronomers has discovered an ultramassive black hole in the foreground galaxy, with a mass estimated to be over 30 billion times that of our Sun. This massive object was detected using gravitational lensing and supercomputer simulations.
Researchers will use satellite data, airborne sensors, and supercomputers to evaluate the impact of tillage practices on corn and soybean yields, greenhouse gas emissions, and soil organic carbon in the Midwest region. The study aims to provide a holistic view of the effects of tillage across a large region.
Researchers have developed a new simulation method to study polarons in 2D materials, which could lead to breakthroughs in OLED TVs and hydrogen fuel production. The study uses quantum mechanical theory and computation to determine the fundamental properties of polarons in 2D materials.
Researchers achieved optical switching of a light signal at attosecond speeds, exceeding data transfer speeds by 1 million times. This breakthrough enables the development of ultrafast optical electronics and could increase data processing speed in long-distance communications.
The Fermi Gamma-ray Space Telescope has captured a dynamic animation of the gamma-ray sky, revealing frenzied activity over nearly 15 years. The data, now publicly available, includes records of source brightness changes and sheds light on blazars and multimessenger astronomy.
A rare quasar triplet formed a massive black hole with a mass of 10 billion solar masses, according to recent simulations. The triple system, composed of three galaxies with supermassive black holes at their centers, is believed to be the progenitor of ultra-massive black holes.
A new European Union project aims to significantly improve Earth system models by incorporating more realistic mesoscale eddy representations. This will provide valuable climate information for national and international assessments, contributing to the IPCC's next Assessment Report.
SpiNNcloud Systems receives EU funding to expand its energy-efficient computing technology, developed from the human brain-inspired SpiNNaker2 supercomputer. The project aims to reduce AI's carbon footprint by creating the most energy-efficient hardware for large-scale applications.
Researchers from Iowa State University and Tufts University are using quantum computing to simulate and study atomic nuclei. They aim to understand the fundamental laws of nature governing nuclear formation in the Big Bang and supernovae.
Researchers have developed a new model of black hole collisions that reveals nonlinear effects in gravitational waves, allowing for more accurate modeling of the behavior. This breakthrough has significant implications for understanding black hole collisions observed by LIGO and testing Einstein's general theory of relativity.
Astronomers have cataloged over 51,863 Lyman-alpha-emitting galaxies, 123,891 star-forming galaxies, and 4,976 active galactic nuclei using HETDEX's spectroscopic data. The survey is a non-targeted, moon-sized survey that collects spectra from 35,000 fiber optic cables, providing a unique dataset for future galaxy mapping.
The EuroHPC Joint Undertaking has renewed support for EuroCC and CASTIEL, two initiatives promoting a consistent standard of expertise in HPC, HPDA, and AI across Europe. The projects have helped strengthen the continent's computational research infrastructure and industrial competitiveness.
A team of researchers is using the Frontera supercomputer to develop medium to long-term fishery forecasts driven by high-resolution coupled climate forecasts. They find that changes to upwelling are predicted to be warmer, not colder, and may impact the sustainability of fisheries in the US and globally.
Researchers have developed a quantum computing architecture that enables directional photon emission, the first step toward extensible quantum interconnects. This breakthrough enables the creation of larger-scale devices by linking multiple processing modules along a common waveguide.
New study reveals marine algae adapt to nutrient-poor ocean conditions, sustaining productivity even in depleted waters. This 'metabolic hack' could impact global ocean productivity and carbon sequestration.
Climate change and land use will cause extinction cascades leading to the loss of 10% of animals and plants by 2050, rising to 27% by 2100. The study highlights coextinctions as a dominant factor in future vertebrate losses.
Using supercomputers and machine learning, researchers created simulations of millions of computer-generated universes to test astrophysical predictions. The study found that supermassive black holes grow in the same way as their host galaxies, revealing a long-elusive relationship.
The Celeritas project aims to address the data tsunami from the upgraded LHC, enabling faster particle collisions and more accurate simulations. With graphics processing units and exascale computing, Celeritas will increase data throughput and validate theories of particle physics.
USTC scientists made a significant breakthrough in ab initio computing simulation of complex metallic heterostructures with 2.5 million atoms. This achievement is expected to be applied in the construction of 2D-materials-based transistors.
The US Department of Energy has awarded $35 million in grants to three joint projects between Nuclear Physics and Advanced Scientific Computing Research programs. These projects aim to optimize software tools for calculations of quantum chromodynamics, which describes the structure of protons and neutrons, using powerful supercomputers...
Researchers discovered a novel interaction between the SARS-CoV-2 Spike protein and human Estrogen Receptor Alpha, which may contribute to severe coagulopathy in COVID-19 patients. The study suggests that modulating ER signaling could mitigate rare vaccine side effects.
Scientists at Argonne National Laboratory have discovered tiny magnetic vortices called skyrmions that could store data in computers, promising 100-1000 times better energy efficiency than current memory. The team used AI and a high-power electron microscope to visualize and study the behavior of these micro-scale magnetic structures.
A team of researchers from Japan Advanced Institute of Science and Technology has discovered thermodynamically stable phases in Y–Ce–H and La–Ce–H systems that exhibit high-temperature superconductivity. Calculations predicted Tc values of up to 173 K, paving the way for the development of more energy-efficient and sustainable societies.
The discussion focused on the impact of ocean acidification, compound events, sea-level rise, and climate injustice, with experts highlighting the need for youth involvement in decision-making. The panel agreed on the urgent need for climate action ahead of COP27, emphasizing cooperation between nations to mitigate the crisis.
Researchers trained a mouse brain model to solve visual tasks, achieving superior robustness and performance compared to traditional neural networks. The model's unique coding properties enable it to cope with errors and unexpected input, making it a promising tool for advances in neuromorphic computing.
Researchers at Paderborn University developed a new algorithm for quantum computing in chemistry, reducing qubit count and increasing parallelisation. This allows for the simulation of larger molecules and improved accuracy despite 'quantum noise'.
A new mechanism has been discovered for the passive transport of biomolecules through the nuclear pore complex, with implications for human diseases. The research team used supercomputing simulations on Frontera and Stampede2 systems to study the kinetics of the nuclear pore transport.
A breakthrough computer model from Chalmers University of Technology reveals the properties of an atomic nucleus, providing insights into the strong force that governs neutron star behavior. The model predicts a surprisingly thin neutron skin, which could lead to increased understanding of heavy element creation in neutron stars.
The Earth System Grid Federation is upgrading its climate projection data system to improve access and curation, with the goal of enabling scientists to make the best guess about the future trajectory of our climate. The new system will provide faster download speeds and enable previously infeasible data analyses.
Researchers at Ohio State University have developed a new machine learning method called next-generation reservoir computing that can learn spatiotemporal chaotic systems in a fraction of the time. The algorithm is more accurate and uses less training data, making it easier to predict complex physical processes like Earth's weather.
A new simulation suggests that gravitational interactions, not over-eating, are responsible for the Milky Way Galaxy's peanut-shaped and nuclear bulges. The study reveals a clear break in star ages between the two types of bulges, indicating when the bar formed.