Among those ranked in the 59th edition of the TOP500 announced at the 2022 International Supercomputing Conference (ISC) in Hamburg, Germany, are standouts conducting powerful research in energy generation, cyber and physical threats, and astrophysics. Plus, Canada’s most powerful supercomputer and Dell’s HPC at the edge.
Eni, HPC5: The Supercomputer Working for Energy
Global energy company Eni was recognized for its HPC5 supercomputer, a powerful and sustainable system used to research new energy sources. Active in 69 countries, Eni works across the value chain—from natural gas and oil to electric and renewable energy generation, traditional and bio-refining and chemicals.
Eni’s HPC5, rated 12 in the TOP500, leverages Dell PowerEdge C4140, Intel® Xeon® Gold 6252, 24C 2.1GHz, NVIDIA V100, and Mellanox® HDR InfiniBand®, with 669,760 total cores, 582,400 accelerator/co-processor cores, 35,450.00 Rmax [TFlop/s], 51,720.76 Rpeak (TFlop/s], and 860.32 HPCG [TFlop/s].
HPC5’s performance level enables sophisticated in-house algorithms to process subsoil data, as well as geophysical and seismic information from around the world. The system, which develops in-depth 3D subsoil models to see up to 15 kilometers underground, led to the discovery of Zohr, the largest known natural gas field in the Mediterranean.
As the energy industry continues to meet high demand for oil and gas, exploration must move to more remote areas including those deeper under the surface of the earth. Eni works to identify new resources and bring them to market as quickly as possible, supported by HPC5’s ability to increase the accuracy of prospecting and speed the launch of production. For optimizing existing resources, Eni uses HPC5 to model, study, and ultimately improve refinement operations.
In initiatives to locate renewable energy resources, such as solar and wind, HPC5 powers plasma and wave research. Eni engineers perform theoretical modeling of photoactive molecules and polymers used for capturing the sun’s energy in solar panels and luminescent solar concentrators. They also build advanced mathematical models, combining meteorological and marine information with information on the behavior of technologies that generate electricity from wave motion.
Texas A&M University, Grace: Cyber-Physical Threat Response
Scientists and engineers at Texas A&M University are conducting research with the power of the Grace supercomputer, named in honor of programming pioneer Grace Hopper whose work in the 1950s led to the COBOL programming language.
Texas A&M’s Grace, rated 231 in the TOP500, leverages Dell PowerEdge R640/R740/R840, Intel® ®Gold 6248R/6248R/6248, 24C 3.0GHz, and InfiniBand® HDR100, with 43,200 total cores, 1,532 accelerator/co-processor cores, 2,638.53 Rmax [TFlop/s], and 4,829.34 Rpeak [TFlop/s]. The University’s Grace-GPU also ranked in the TOP500, coming in at 365.
The Grace system handles nearly 20 times the processing of its predecessor and supports more than 2,600 researchers in groundbreaking studies on drug design, materials science, geosciences, fluid dynamics, biomedical applications, biophysics, genetics, quantum computing, population informatics, and autonomous vehicles.
One such project, led by Dr. Katherine Davis, assistant professor in the Department of Electrical and Computer Engineering and her research team, is looking at protecting critical energy infrastructure. Specifically, they are interested in electric utility response to cyber and physical threats, and they are working to develop an algorithm that can be used as a tested, trusted safeguard. Davis and her team are using HPC-driven data science and AI to model solutions whereby energy management systems would protect themselves with semi-automated, real-time data analysis.
The aim of their project is to develop “a scalable, physics-informed and AI-enabled cyber-physical intrusion response solution for electric power utilities,” according to news from the Texas A&M Engineering Experiment Station. Davis and her team are building on an earlier project, known as Cyber Physical Resilient Energy Systems, that focused on a secure end-to-end system that could provide a modeling foundation for next-generation energy management.
Texas Advanced Computing Center, Frontera: A Beastly Black Hole
The Frontera supercomputer, funded by the National Science Foundation (NSF), powers discoveries for hundreds of scientists and engineers including the first all-atom simulations of the COVID-19 protein spike, the most realistic simulations of supercell thunderstorms, and the largest cosmological simulation performed at high resolution. Known as the most powerful supercomputer in academia, Frontera is hosted by the Texas Advanced Computing Center(TACC) at the University of Texas, Austin.
Frontera earned the 16th spot in the TOP500, and TACC’s Stampede2 supercomputer was rated at 47. Frontera leverages Dell PowerEdge C6420, Intel® Xeon® Platinum 8280, 28C 2.7GHz, and Mellanox® InfiniBand® HDR with 448,448 total cores, 23,516.40 Rmax [TFlop/s], and 38,745.91 Rpeak [TFlop/s].
Recently, the breaking news spotlight was on Frontera for its role in producing the first image of Sagittarius A*, the supermassive black hole at the center of the Milky Way galaxy. A global research team known as the Event Horizon Telescope Collaboration produced the image—with observation and data analysis experts working alongside others who developed computer models based on the known physics of black holes.
“The accreting flow around the galactic center black hole is complex,” Chi-kwan (CK) Chan, with the Steward Observatory and Department of Astronomy and Data Science Institute of the University of Arizona, said in a press release. “Advanced computing systems like Frontera allow us to use cutting-edge computational science methods to model these complex systems, study the boundaries between order and chaos, and understand what we are seeing with the Event Horizon Telescope.”
Chan noted that the HPC-powered models provide a “deep understanding on how Sagittarius A* is shaped by its plasma environment.” Now, researchers are using Frontera to ask and answer new questions: How will Sagittarius A* change over time? Can we probe the discrepancies between Sagittarius A* computer models and observations?
Canada’s Most Powerful Supercomputer: The 5.9 Petaflops Narval
Coming in at 92 in the TOP500 was Calcul Québec’s Narval (GPU), a supercomputer dedicated to scientific research in Québec and Canada. Calcul Québec, a group dedicated to sharing high-performance computing infrastructure, is a regional partner of Compute Canada, a non-profit organization that works to coordinate and facilitate the delivery of Canada’s national advanced research computing platform.
Narval is a Dell PowerEdge system equipped with 80,000 CPU cores, 422 TB of memory, and 25 PB of disk storage, as well as 632 A100 graphics processing units (GPUs) from NVIDIA. Narval is housed alongside another one of Calcul Québec’s supercomputers, Beluga, which was rated 312 in the TOP500.
Pierre Fitzgibbon, Minister of the Economy and Innovation and Minister responsible for Regional Economic Development, said in a press release: “With the availability of Narval, Calcul Québec is helping to make Québec a leading scientific center for research, training, technology transfer, and the creation of value-added products. This supercomputer will encourage the realization of new research projects, generating major spin-offs in strategic sectors of Quebec’s economy.”
The Rattler: Dell’s Own HPC at the Edge
Earning the rank of 254 in the TOP500 is the Rattler supercomputer from Dell Technologies. The Rattler system is one of three powerful HPC clusters at the Dell Technologies HPC & AI Innovation Lab, a world-renowned center in Austin, Texas. Dell engineers and subject matter experts work with customers at the HPC & AI Innovation Lab to design IT-specific workload solutions, test emerging technologies against existing applications, and develop best practices.
Dell’s Rattler leverages Dell PowerEdge servers with NVIDIA A100 80GB, and InfiniBand® HDR200 dual rail, with 26,880 total cores, 20,736 accelerator/co-processor cores, 2,531 Rmax [TFlop/s], 3,940.61 Rpeak [TFlop/s], 999,936 Nmax, and 43.25 HPCG [TFlop/s].
A recent Dell Technologies blog stated, “This story gets even better when you consider that part of the Rattler cluster is in a Modular Data Center in the parking lot of a Dell Technologies building in Austin, Texas. That makes Rattler truly HPC at the physical edge.”
Intel® Technologies Move Analytics Forward
Data analytics is the key to unlocking the most value you can extract from data across your organization. To create a productive, cost-effective analytics strategy that gets results, you need high performance hardware that’s optimized to work with the software you use.
Modern data analytics spans a range of technologies, from dedicated analytics platforms and databases to deep learning and artificial intelligence (AI). Just starting out with analytics? Ready to evolve your analytics strategy or improve your data quality? There’s always room to grow, and Intel is ready to help. With a deep ecosystem of analytics technologies and partners, Intel accelerates the efforts of data scientists, analysts, and developers in every industry. Find out more about Intel advanced analytics.