Intel Xeon Scalable processors drive advanced research in world's fastest academic supercomputer
Thierry Pellegrino (from left), Dell vice president of Business Strategy & HPC solutions, Dan Stanzione, executive director of the Texas Advanced Computing Center, and Trish Damkroger, Intel vice president and general manager of the Extreme Computing Organization, display a plaque commemorating Frontera on September 3, 2019. The supercomputer at the Texas Advanced Computing Center at the University of Texas at Austin will enable breakthrough research across a number of fields. Credit: Texas Advanced Computing Center.3 Sep 2019 Santa Clara - Intel, the Texas Advanced Computing Center (TACC) at the University of Texas at Austin, the National Science Foundation (NSF), Dell Technologies and other science and technology partners have gathered to unveil Frontera, the fifth most powerful supercomputer in the world and the fastest in academia. Based on 2nd generation Intel Xeon Scalable processors and featuring Intel Optane DC persistent memory, the Frontera system is poised to accelerate scientific research and innovation.
"The Frontera system will provide researchers computational and artificial intelligence capabilities that have not existed before for academic research. With Intel technology, this new supercomputer opens up new possibilities in science and engineering to advance research including cosmic understanding, medical cures and energy needs", stated Trish Damkroger, Intel vice president and general manager, Extreme Computing Organization.
As the world's fastest academic supercomputer, Frontera will enable breakthrough research across a number of fields, including astronomy, medicine, artificial intelligence (AI), quantum mechanics and mechanical engineering.
Early projects being run on Frontera include:
Understanding the influence of distant stars: Manuela Campanelli, professor of astrophysics at Rochester Institute of Technology and director for the Center for Computational Relativity and Gravitation, is using Frontera to develop a simulation to amplify our understanding of gravitational waves. The goal is to explain the origin of the powerful energy bursts that are emitted during a neutron star merger, including the types of electromagnetic signals emitted. Frontera enables Manuela Campanelli and her team to perform complex simulations at a speed two or more times faster than what is possible on any local supercomputer.
Diagnosis and treatment of gliomas: Professor George Biros of UT Austin is using Frontera to build complex bio-physical models of brain tumour development to more effectively diagnose and treat gliomas, a type of brain tumor. Fronteras state-of-the-art system enables automated medical image processing to detect the extent of cancers beyond the main tumor growth, which must be removed during surgery to prevent the cancer from returning.
Teaching neural networks quantum chemistry: Olexandr Isayev, assistant professor at University of North Carolina at Chapel Hill, is using Frontera to train a neural network that accurately describes the force fields and potential energy of molecules based on their 3D structures. The Frontera system enables Isayev to scale up his research, reaching a record of 3 million calculations in 24 hours. His work has many potential applications, but among the most impactful is drug discovery - finding new molecules that can interact with specific proteins to treat and cure diseases.
Eradicating emerging viruses: Peter Kasson, associate professor of molecular physiology and of biomedical engineering at the University of Virginia, is studying emerging viruses and guiding the development of new antiviral therapies. To study the mechanisms of these viruses, Kasson and his team combine microscopic research with computer models of the virus, built one atom at a time, and then simulate the mechanics of how the atoms would interact. Early work on Frontera is already enabling simulations that are two or three times faster than on prior supercomputers.
Building a sunnier energy future: Ganesh Balasubramanian, assistant professor of mechanical engineering and mechanics at Lehigh University, was among the early users of Frontera, studying the dynamics of organic photovoltaic materials. Actively collaborating with experimentalists, he is working to develop efficient ways to create next-generation flexible solar photovoltaics that can exceed the energy-producing potential of today's devices.
First announced in August 2018, Frontera was built in early 2019 and earned the no. 5 spot on the TOP500 list of the world's most advanced supercomputers in June 2019. Featuring Dell EMC PowerEdge servers with 2nd Gen Intel Xeon Scalable processors, Frontera's 8008 compute nodes delivers peak performance of 38.75 petaflops. Frontera also features Intel Optane DC persistent memory to address the performance and memory capacity requirements of the diverse workloads run in the system, including scientific modelling and simulation, Big Data and artificial intelligence.
Intel and TACC have been collaborating on groundbreaking supercomputing projects for more than a decade, with the goal of developing advanced systems to enable scientific breakthroughs for the research community. The Frontera supercomputer continues the close collaboration between TACC, Dell Technologies and Intel to drive high-performance computing innovations with unprecedented performance, productivity and efficiency.