Source: DOE | Interview | March 30, 2020
Last week, President Trump announced the COVID-19 High Performance Computing Consortium, a unique public-private effort spearheaded by the White House Office of Science and Technology Policy (OSTP), the U.S. Department of Energy (DOE), and IBM – along with a growing group of government, industry, and academic leaders – to unleash the power of America’s supercomputing resources to combat COVID-19.
Under Secretary for Science Paul Dabbar is leading the Consortium’s efforts at DOE, and he described them in more detail via a (virtual) interview below:
Could you talk a bit more about the Consortium: What makes it unique? How will it speed our efforts to stop the virus?
To accelerate the fight against COVID-19, the White House, in partnership with IBM and the Department of Energy, announced on March 22nd the creation of the COVID-19 High-Performance Computing Consortium. This public-private partnership includes the biggest players in advanced computing from government, industry, and academia. At launch, the consortium includes five DOE laboratories, industry leaders like IBM, Microsoft, Google, and Amazon, and preeminent U.S. universities like MIT, RPI, and UC San Diego. And within a week, we’ve already received more than a dozen requests from other organizations to join the consortium.
Never before has a group of public and private entities–federal government, industry, and academic leaders–organized so quickly and made so much time available on so many of the world’s most powerful computers at no cost to support the community of researchers tackling COVID-19’s big problems. And it’s essential to ensure we can respond as quickly as the crisis merits. These powerful computers will enable researchers to quickly understand how the disease may spread in communities as conditions change on the ground. They will allow researchers to rapidly identify targets on the virus and, within humans, to more quickly screen potential pharmaceutical treatments and develop a future vaccine. And they will allow researchers to predict how the virus may change as it spreads through the population.
How will the Consortium work – how will projects be selected and computing time be allocated? Will it focus on particular areas?
An expert panel comprising top scientists and computing researchers will assess the public health benefit of the proposed work, with emphasis on projects that can ensure rapid results. Once a project is selected for support, a Consortium member will assume responsibility and work with the proposal team to identify the process for access and to discuss any specific terms and conditions that will apply to the offered access.
How does a researcher submit a proposal for consideration?
Researchers are encouraged to submit COVID-19 related research proposals to the Consortium via an online portal (https://www.xsede.org/covid19-hpc-consortium), which will then be reviewed for matching with computing resources from one of the partner institutions.
Which DOE National Labs are involved? Why were they selected? What makes their capabilities so important in the fight?
Currently, six DOE National Labs are involved in the Consortium: Argonne, Lawrence Berkeley, Lawrence Livermore, Los Alamos, Oak Ridge, and Sandia National Laboratories.
The DOE National Laboratories selected have world-leading supercomputing capabilities that are uniquely suited to performing complex simulations, which are increasingly employing machine learning and artificial intelligence as a critical component of the model. It is this capability that DOE will use to address the COVID-19 challenges.
DOE is already using AI to search for drugs that might be effective on the virus, to identify those proteins in both the virus and in the host that will be good drug targets, to look for important mutations related to vaccine development and drug targeting, and to highlight effective strategies to minimize the load on the health care system and maximize the protection to vulnerable groups. This effort is enabled by work our labs have been doing for several years leveraging AI in the fight against cancer. The lessons learned from that work have been quickly transitioned to support similar efforts for COVID-19.
The combination of DOE supercomputers with the increasing expertise at the DOE laboratories for incorporating AI into scientific research can be of great service in this battle. Working with pharmaceutical and diagnostics companies, DOE is accelerating the discovery of promising treatments. Working with policymakers, we are devising effective ways to manage the course of infection and deploy much-needed resources strategically, to the locations where they will do the most good.
Beyond the Consortium, how else are DOE’s National Labs contributing to our efforts against COVID-19?
The DOE laboratory complex has many core capabilities that can be applied to addressing the threats posed by COVID-19. In addition to the computational modeling and simulation described above, our labs are using their suite of X-ray light sources and neutron sources to rapidly make molecular structure determinations of key proteins from SARS CoV-2, the virus that causes COVID-19. These structural data are an essential input for many of the computational studies already underway. Our labs will continue working with collaborators from across the country to leverage these world-leading facilities towards better understanding this virus.
DOE’s labs are also helping to tackle supply chain bottlenecks. A new 15-lab task group, guided by input from both public and private stakeholders, has identified three health care supply chain bottlenecks and has assigned teams to rapidly address the issues, which include surgical masks and face shields, ventilator systems, and consumables (e.g., swabs, test kit components) used in COVID-19 testing. Efforts are currently underway to use additive manufacturing to rapidly prototype test articles. Once the prototypes are validated, the task force will assess production considerations, including design and manufacture of tooling, material selection, and supply chain engagement. The primary focus will be on the development of dies and molds that can be rapidly and broadly distributed to private companies experienced in the manufacturing of health supplies, using protocols that meet regulated health care standards.