While supercomputers have played a significant role in advancing life sciences research, the COVID-19 pandemic has crystalized just how important they are to dealing with this crisis. In an unprecedented move, the COVID-19 HPC Consortium has launched to provide access to the world’s most powerful high-performance computing resources in support of COVID-19 research.
The consortium is a public-private collaboration initiated by the White House Office of Science and Technology Policy to ensure critical COVID-19 research has the compute cycles it needs. Additionally, the effort has HPC centers, research teams, and the open-source community partnering to optimize performance for vital tools and applications to enable faster progress.
Bring Supercomputing Power to Bear on Research
Fighting COVID-19 will require extensive research in bioinformatics, epidemiology, and molecular modeling to understand the threat and develop strategies to address it.
Consortium members and affiliates manage a range of computing capabilities: from small clusters to some of the world’s largest supercomputers. They offer not only computational resources, but also software, services, and deep technical expertise to help COVID-19 researchers execute complex computational research programs. Collectively, the consortium offers access to 485 petaflops, 5 million CPUs, and 50,000 GPUs.
A large part of this collective power is delivered via supercomputers based on Intel technology. The group includes the world’s top-performing supercomputing centers, such as the Texas Advanced Computer Center (TACC) at The University of Texas, Department of Energy’s Argonne National Laboratory, the Pittsburgh Supercomputing Center, and more. Additionally, Germany’s Leibniz Supercomputing Centre and the King Abdullah University of Science and Technology in Saudi Arabia have invited coronavirus-related research teams to apply for time on their Intel processor-based HPC platforms.
An example that illustrates the growing compute requirements and how a suitably selected HPC solution can help is the BIH Digital Health Center of the Berlin Institute of Health (BIH) and its advanced COVID-19 cell sequencing research and how critical technology is in delivering timely results within the medical industry.
Samples of other work being carried out includes:
At the molecular level, research teams are building detailed digital models of the novel coronavirus and running physics-based simulations to explore how it works and what aspects of its molecular structure offer potential targets for vaccines and treatments. One such team at the University of California San Diego (UCSD) is constructing a complete model of the virus’s exterior shell—the part we’re trying to dissolve when we wash our hands for 20 seconds. They’re also planning to simulate the atom-to-atom interactions of the 200 million atoms that make up the shell.
At the macro level, scientists and policy analysts want to better understand how COVID-19 is spreading and what containment strategies can best minimize disease transmission and economic disruption. Epidemiological simulations empower them with data for developing science-based answers to these questions. In one of the largest COVID-19 related epidemiological studies, an 80-person team at the University of Virginia Biocomplexity Institute and partner institutions have built a detailed “digital twin” that models the entire United States population. They use this model to simulate how the U.S. would respond to possible pandemic-related actions to inform answers to the most pressing “what if” questions. The model runs each night at the Pittsburgh Supercomputing Center, and results are shared with federal and state agencies.
Underlying Technology is the Key to Success
Most of this COVID-19-specific research conducted at these consortium facilities was simply not possible just a few years ago. That is a testament to the incorporation of the latest technologies into the systems at those centers.
Specifically, new Intel technology has allowed the most powerful supercomputer systems to scale performance and increase workload capacity. For example, many of the most powerful systems are built with 2nd generation Intel® Xeon® Scalable processors and featuring Intel Optane™ D.C. persistent memory. Such systems poised to accelerate scientific research and innovation.
In addition, Intel is providing HPC /AI and HLS subject matter experts and engineers to collaborate on COVID-19 code enhancements to benefit the community. Intel also will provide licenses for HPC software development tools for the research programs selected by the COVID-19 HPC Consortium. The integrated tool suites include Intel’s C++ and Fortran Compilers, performance libraries, and performance-analysis tools.
Taking advantage of Intel technologies, scientists are applying their expertise and passion for advancing the world’s understanding of the disease, its causes, and its impacts. Industry leaders are collaborating with government agencies and universities to support research initiatives that can help lead to cures and vaccines and provide a basis for data-informed policies.
In doing so, they’re providing a model of how to solve the biggest problems. They’re advancing their algorithms and software in ways that are crucial for understanding COVID-19—and can also help deliver faster, deeper insights into the science behind other global challenges and opportunities.
Keep up with Intel HPC developments >> https://www.hpcwire.com/solution_channel/intel/
Learn how scientists and researchers are investigating COVID-19 with powerful Intel technologies >> https://www.intel.com/content/www/us/en/corporate-responsibility/covid-19-supercomputing-article.html