BLOOMINGTON, Ind.—Indiana University’s Big Red 200 supercomputer is starting to help IU student and faculty researchers make discoveries in artificial intelligence (AI), medicine, cybersecurity, and hundreds of other disciplines. True to the “HPC Accelerates” theme of this year’s Supercomputing Conference, Big Red 200 will create new capabilities that enable researchers to analyze data and perform calculations at a rate 6 times faster than its predecessor, Big Red II, and 400 times faster than the original Big Red supercomputer installed at IU in 2006.
With greater than 7 quadrillion floating-point operations per second, Big Red 200 can do in a single second what it would take 28 years for everyone in the state of Indiana to do, performing one calculation per second without stopping.
“The state-of-the art simulations enabled by the computational power of Big Red 200 will provide the theoretical framework necessary for the interpretation of observations made with the Hubble Space Telescope and the new James Webb Space Telescope, which launched in December,” said Enrico Vesperini, an associate professor in the Department of Astronomy. Vesperini and his research group will use Big Red 200 to run large surveys of numerical simulations aimed at exploring globular star clusters—very dense stellar systems that are among the oldest in the universe. “Globular star clusters play a key role in the study of many fundamental questions in astronomy,” Vesperini added.
“Strategic investment in our technology infrastructure is vital when it comes to innovating and performing research that feeds into the research and education engine,” said Rob Lowden, IU vice president for information technology and chief information officer. “By making Big Red 200 available to Indiana’s researchers, we’re drastically increasing their ability to pursue projects and goals that may have been previously perceived as impractical—or completely unattainable. We expect the research done on Big Red 200 will lead to new advancements in medicine, cybersecurity, climate research, and so much more and directly contribute to student success.”
Big Red 200 is accelerated by 256 NVIDIA A100 Tensor Core GPUs and is the first of Hewlett Packard Enterprise’s (HPE’s) revolutionary new Cray EX supercomputers installed at a U.S. university. The family of systems from HPE are being deployed at a larger scale as part of the Exascale Computing Project (ECP) led by the U.S. Department of Energy. ECP aims to develop the world’s fastest supercomputers with speeds in excess of 1018 calculations per second. The United States’ first exascale computer, Frontier, will be installed at Oak Ridge National Laboratory later this year. Other core DOE labs partnering in the ECP are Argonne National Laboratory, Berkeley Lab, Lawrence Livermore National Laboratory, Los Alamos National Laboratory, and Sandia National Laboratories.
“Big Red 200 is a breakthrough supercomputer with capabilities to allow researchers to push the boundaries of scientific exploration as we enter the era of exascale AI,” said Cheryl Martin, director of higher education and research at NVIDIA. “NVIDIA GPU-accelerated computing provides incredible performance and flexibility to the wide range of HPC and AI workloads Big Red 200 will tackle, enabling scientific innovation that will deliver countless benefits for society.”
Big Red 200 will be IU’s premier system supporting hundreds of millions of dollars in grant-funded research annually. With Big Red 200, IU’s research community will be able to do research at a scale and speed that was previously impossible, empowering student success and helping IU to continue as a cutting-edge leader in research and discovery. Big Red 200 accounts will be available to IU researchers beginning the week of April 4, 2022.