COVID-19 is an unprecedented threat. With 1.2 million confirmed cases and 72,000 deaths in only a few months’ time, the disease has quickly spiraled into a global crisis that doesn’t show any sign of letting up. To combat it, we’ll need unprecedented scientific effort and cooperation.
Thankfully, high-performance computing (HPC) communities are ready for a fight. The COVID-19 HPC Consortium, led by IBM, the White House Office of Science and Technology Policy, and the Department of Energy (DoE) brings supercomputing facilities together to pool resources.
We asked DoE Under Secretary of Science Paul Dabbar what kinds of projects the Consortium is looking for, why HPC is so important right now, and what it’s like to face such an enormous challenge.
How was the COVID-19 HPC Consortium formed?
It primarily started off with a conversation between IBM, the White House, and then the DoE, talking about the HPC resources that not only we had, but the other HPC resources around. And obviously, the DoE has the biggest supercomputers in the world (Including #1-ranked Summit, at Oak Ridge National Laboratory in Tennessee –Eds.). IBM is a major manufacturer of high-performance computers, including some of ours. But they have their own groups that they work with.
This is a problem for humankind. We're here to volunteer, we have resources, we want to help.
And so, they had their community and we had our community. We overlapped but obviously it wasn't 100%. IBM brought up that there were other resources out there that would be exciting to try to support this research. So, the White House, IBM, and the DoE got together and said “Well, why don't we form this consortium to try to bring in resources.” We also want to jointly solicit research.
Why is HPC a valuable tool in the fight against COVID-19?
There’s a number of things around bioinformatics, around epidemiology, around molecule design that high-performance computers can work on. That's kind of a lot of the high end of what HPC's capacity is at the Department of Energy National Laboratories.
The big areas that we would do modeling around are materials, chemistry, and biology, and also physics. And that's a very key part of this, which is taking a look at COVID-19 and trying to compare that to what we generally have from a bioinformatics point of view, including prior coronaviruses.
Also, trying to identify which particular small molecules out there that might be able to go after proteins in COVID-19, and either provide therapies or vaccines. Modeling the expansion of the disease is certainly a potential there that we also have.
When we last spoke in May 2018, the DoE was pushing to improve US supercomputing resources. You said then that the DoE’s focus on HPC was “providing the impetus for innovation within supercomputing.” How does that resonate with you now?
I think it resonates an awful lot with me personally—and hopefully with many others. The National Labs, as former Secretary Perry would say, are the crown jewels for research in the United States.
There are 17 National Labs and 60,000 people, making it the largest basic research entity in the world. But also we build facilities that are useful for not only ourselves but also as user facilities, which means that other people can use them for research.
And so, within supercomputing, there's applications in chemistry, there's applications in materials, there's applications in biology and physics. They're flexible to be allowed for potentially different applications. You never know what the most important issue that may come up will be. These flexible facilities, on behalf of the country and the world, are pivoting very strongly in this particular direction right now.
What strengths do academic institutions bring to the Consortium?
Clearly, academic institutions bring researchers. Some certainly bring HPC capacity, like University of Texas at Austin and University of Illinois, and others. But clearly, a strength of academia is also the research breadth. There are universities all over the country, and they've been doing research in these areas for a long time. Across this country, researchers from academia want to jump in, and they have great ideas on how to attack this.
Across this country, researchers from academia want to jump in, and they have great ideas on how to attack this.
The federal government is the biggest funder of grants in the country for researchers. And so we're very used to working with the broad research community on all sorts of different topics. We know them well, we know how to fund them, we know how to communicate with them, we speak the same language between needs and research.
It’s great having this HPC Consortium to say, “Hey, we have even more resources available than ever before because of the public-private partnership aspect, please bring us your ideas and we'll screen them quickly as a consortium and give you access quickly.”
We’re very excited about academia. There are already many proposals that are being submitted and are already being screened, and are already being allocated. We're already off and running.
You’ve previously worked in the public, private, and academic sectors. The COVID-19 HPC Consortium is uniting people from all three of these areas. What’s it like watching them all come together?
I think it's particularly exciting in this consortium that a large group of tech companies decided to volunteer and to support this critical time in public health for the world. I think a lot of us in the administration right now come from both a public and a private background.
When the idea came up of having the likes of IBM, Microsoft, Google, Amazon Web Service, and HPE/Cray all wanting to contribute, to try to help drive the solutions for the COVID crisis, our immediate reaction was, "Yes, and how do we organize that?"
It's very exciting to see our fellow Americans start contributing whatever they can.
It certainly comes from a mindset of “How do we work together?” IBM is one of our major partners in building supercomputers at our labs. So, we're very well-aligned day-to-day. But I think it is truly a testament to the country and this particular segment of the economy to see Microsoft, Google, Amazon, HPE, and IBM all coming together and volunteering their resources and their people.
Seeing the tech companies all come together and say, “This is a problem for humankind. We're here to volunteer, we have resources, we want to help.” I find it very exciting that they volunteered, and I think we're off to a good start in this public-private partnership.
Many people believe this to be one of the greatest challenges the world has ever faced. What’s it like being part of the fight against COVID-19?
The Department is a very large organization and there's been many things going on. We have approximately 115,000 employees either as direct federal employees or working with the National Labs in a variety of our sites. So, it's a very large organization, and a large part of what we do is focused around operations.
It's key for us to make certain that our employees are healthy during this time of crisis. That is quite a bit of work in the managerial and leadership piece of that large organization. For myself and the secretary and the others in leadership, we're very, very focused on that public health aspect.
Another aspect is that we still have a mission, right? And within the realm of merging the occupational health of our partners is the question of what do we prioritize with that. The three big missions of the department are still extremely active, even in this time of crisis.
We obviously serve national defense and we're a very key aspect of it, which is strategic weapons that we still need to maintain going through this process. The second part is clearly the energy mission. There’s a very challenging situation in the energy industry amongst other areas of the economy. But we do run the Energy Department dealing with what's going on with market share, conflicts going on between other countries at the moment. So, that's quite active and we are having to have to manage that at the same time.
Finally, we run the National Laboratory Complex and so that's what we were talking about, which is how do we keep open with the occupational health challenges that we face. Keeping the key people of resources and supercomputing healthy while also keeping the most critical infrastructure up and running, which is the supercomputing capacity and the energy capacity.
I think our colleagues at the National Labs and our colleagues within this consortium are very excited about trying to make a difference. It's really exciting when I talk to people, even outside of those areas. They have a smaller footprint on the potential, but they want to contribute as well. It's very exciting to see our fellow Americans start contributing whatever they can.
It's important for us to let everyone know that there are people working hard to contribute— fellow Americans and globally—toward this topic. There's lots of people working hard, the COVID-19 HPC Consortium is a small subset. We hope we can contribute to making a difference.
If you have research you would like to propose to the COVID-19 HPC Consortium, click here and follow the instructions.