Fighting Against COVID-19 With High-Performance Computing

Photo credit: iStockphoto/ipopba

Last week, the federal government announced the launch of its COVID-19 High Performance Computing Consortium, a unique public/private industry initiative to support researchers across the globe to harness and leverage supercomputers in the U.S. The White House spearheaded this move along with the US Department of Energy (DoE) and IBM. The consortium includes 16 entities that come from the U.S. government and academic leaders, all of whom have volunteered free compute time and resources on their high-performance machines. In total, the combined computing resources represent 16 systems with more than 330 petaflops, 775,000 central processing unit cores, and 34,000 graphics processing units, meaning it can perform around 330 trillion floating-point operations per second. A few of these members and its capabilities are:

  • IBM and Oak Ridge National Laboratory and the University of Tennessee. The two groups are using IBM’s Summit, the world’s most powerful and smartest supercomputer, to identify 77 small-molecule drug compounds that may aid in fighting against COVID-19. The researchers performed simulations on Summit of more than 8,000 compounds to screen those likely to bind to the main “spike” protein of COVID and therefore rendering it unable to infect host cells (i.e., humans).
  • Rensselaer Polytechnic Institute (RPI). The university is making available AiMOS to the high-performance-computing (HPC) consortium. This contribution is significant, as the supercomputer is the most powerful among any private university in the US with a sustained processing rate of eight petaflops (equivalent to 8 quadrillion calculations per second). Currently, AiMOS is assisting the consortium to answer questions about COVID-19’s spread over time and its effects, explore potential vaccine and treatments, and analyze materials that could be used for personal protective equipment for medical workers.

While the consortium underscores HPC’s relevance in solving today’s problems, it isn’t the first time the technology has been used in a science and medical setting. In fact, drug discovery and DNA sequencing were some of the earliest and most common HPC use cases. Today, this focus is shifting to help the global response to COVID-19. Other examples include:

  • Amazon Web Services. The company recently launched a USD 20 million initiative to fight COVID-19 by offering research institutions and companies technical support and promotional credits for use of its programs to advance research on diagnosis, treatment, and vaccine studies.
  • National Supercomputing Center of Tianjin. The laboratory used its system, Tianhe-1A. The AI-assisted analysis is used to determine whether sick patients were infected with COVID-19 or had common viral pneumonia.
  • Gauss Centre for Supercomputing. An alliance between Germany’s three major supercomputing centers is supporting COVID-19 efforts by expediting access to HPC infrastructure at the High Performance Computing Center Stuttgart (HLRS), Jülich Supercomputing Centre (JSC), and Leibniz Supercomputing Centre (LRZ). The three have committed to minimizing hurdles for research on the molecular-level study of the virus, vaccine and therapeutics development, epidemiological research, and disease spread forecasting.
  • [email protected] The largest crowdsourced supercomputing program in the world kick-started an initiative to uncover the major infection device of COVID-19, its largest spike protein. Since its February start, the COVID-19 initiative has attracted roughly 400,000 volunteers.

We are using these case studies as a launching point for our HPC research. In the future, we plan on releasing a new report on high-performance computing that highlights new use cases and implementation that have emerged in the past couple of years. 

The original article by Tracy Woo, senior analyst serving Infrastructure and Operations professionals at Forrester, is here. The views and opinions expressed in this article are those of the author and do not necessarily reflect those of CDOTrends. Photo credit: iStockphoto/ipopba