The U.S. Department of Energy’s High Performance Data Facility hub will make its home in Newport News, as announced by the DOE Office of Science last week. A longtime partner of William & Mary, the Thomas Jefferson National Accelerator Facility (Jefferson Lab or JLab) will lead the new $300+ million hub specializing in advanced infrastructure for data-intensive science, accelerating scientific research and discovery.
William & Mary strongly supported the Jefferson Lab’s proposal to lead the new facility. W&M News discussed the importance of this partnership with Dennis Manos, CSX Professor of Applied Science and Physics and vice provost for research.
The interview has been edited for length and clarity.
Q: How has the relationship between William & Mary and Jefferson Lab evolved throughout the years?
A: Our relationship with Jefferson Lab goes back to its earliest inception. Historically, the state worked through William & Mary to possess a property that once served as the Space Radiation Effects Laboratory for NASA, where accelerators had been in use earlier. The SREL was on the site that Jefferson Lab currently occupies. At that time, W&M physics faculty members Robert Siegel, Hans von Baeyer and Franz Gross, working with James McCarthy of University of Virginia, were instrumental in putting together a winning proposal to the Department of Energy.
After a lot of wrangling, by 1986 this effort led to the 4 GeV electron ring now known as the Continuous Electron Beam Accelerator Facility. One very novel feature of the later stages of that proposal was the idea that it should use superconducting cavities to replace what until that time had been normal-conducting copper components. In that decision, a lot of very interesting accelerator-related physics, engineering and material science emerged, allowing Jefferson Lab to make contributions not only in nuclear physics, but in related areas so long as the related work could be justified as serving the nuclear physics mission.
Q: How will the creation of the HPDF further develop this link, and what opportunities will it open?
A: Since the time I came to Virginia, we have been working on a variety of topics to help Jefferson Lab to relax the constraints associated with being a single-purpose laboratory. In that regard, building the new High Performance Data Facility will create that change very quickly, permitting it to become a lead laboratory for the national lab system, and allowing JLab scientists to work on a very wide range of technical topics of great importance. That is the administrative importance of this hub.
The scientific, engineering and technology content of the hub will be truly amazing. It will create a platform to develop physical resources to allow exabyte computing (equal to, or greater than a billion-billion arithmetic operations each second). There will also be work on non-traditional computers whose hardware is based on “entangled qubits” to introduce quantum computing methods and quantum information systems for communications and security applications.
We hope ultimately to see great advances in information storage systems capable of rapidly storing more than yottabytes of information safely and securely.
What this means is that 10 years from now, JLab will be taking on problems that you wouldn’t think of taking on now. Using a combination of traditional and quantum information systems, you will be able to take on problems in chaotic dynamics, or immensely complicated ecological or environmental problems, very large network problems, and more. This is all very, very different from what we’ve been doing so far. It’s going to be an exciting time.
Q: How does this further W&M’s goals under the data and careers initiatives of the Vision 2026 strategic plan? And how does it intersect with the university’s exploration of a new academic unit in computing, data science and applied science?
A: For the last five years, working with the provost and the president and many other people, William & Mary technologists have been putting together the necessary path to bring together physics, data science, computer science and applied science.
A detailed proposal for this new unit, which will offer every phase of education, including undergraduate and Ph.D. degrees, is almost ready to go forward to the Board of Visitors and the State Council on Higher Education in Virginia. Hopefully, once approved, we can bring it immediately to bear on these interesting problems.
If you look at the W&M strategic plan, Vision 2026, you’ll find that we intend to bring the best version of high-touch liberal arts education, with both breadth and depth, to all the students who come to our school. And this hub will be a perfect example of tools to bring the full weight of technology to serve our liberal arts programs because every department and program has the capacity to bring something important to —and take something important from — the data science and engineering methods available in this enormously powerful center.
Some problems are born so hard that we call them wicked, like eliminating inequality, poverty and violent conflict or the problems of sustaining democracy itself. We do not know how to prove that these problems even have solutions, but our faculty seek to use the most powerful technology to examine the development of better predictive models of large-scale organizations that may guide potential directions for policy, law or institutional operations to at least relieve some of the pressures we are experiencing.
The technology that is yet-to-come, like quantum information systems, may help us approach problems that do not map onto simple algorithms. In the coming hub, we hope to seek solutions using machines built from qubits. Such computers don’t yet exist at a large enough scale to be tried, and there appear to be many problems in creating large machines that might be able to run long enough to reach an answer. The quest to develop them is worth the effort though, since they just might allow us to predict the future a bit better, by simulating it, to include all of the various branching probabilities.
“Prediction” is one of the hallmarks of advanced science; good prediction allows better “design.” Good prediction also allows better control. Good design and control are the hallmark of advanced engineering technology, on which good economies depend. So, it is in the W&M strategic plan to develop data methods to allow us to predict a bit more accurately, and a bit farther out in time, to allow us to control the changes we know must occur. All this serves our W&M internal economy, as it serves the Commonwealth we belong to.
Q: What does the creation of this hub this mean for students from different fields? What kind of interest have we been seeing from them?
A: Over the past few years, we have seen 2D and 3D artists come over to our Makerspaces to use our scanning electron microscopes to visualize things that are too small to see in order to render representations of them. Creative artists use whatever tools they can to bring impressions of the world into their own head, to interpret it, redefine it and refine it to create a perception they can share with other people. That’s the very function of the tool we call a language.
Exascale computers provide yet another such tool, that people in sociology, political science, art, music, religion might be able to use. Such sharing of modeled human experience is not intended to be enjoyed only by the people who create the technology.
The coming data hub will evolve to all good uses in ways that are independent of the mathematics, computer science, applied science and physics, and also independent of the communication device manufacturing.
So, people encountering technology will always find their own purpose, their own enjoyment and their own reactions to it. This is why I am confident that William & Mary is better positioned than most other schools to take full advantage of this hub. Yes, like others, W&M has technologists, but our technologists genuinely appreciate the work of W&M non-technologists.
Q: Can you give a few examples of the Jefferson Lab projects our students and faculty are currently involved in?
A: There are very many we could mention. Jefferson Lab has been terribly important in providing opportunities for summer, part-time and full-time research and intern work. So far, JLab engagements have been with graduate students or upper-division undergraduate students, working on accelerator technology or nuclear physics problems. We have done very interesting student work on detector development. One example is the development of high-precision, position-sensitive gamma-ray detectors, which led to medical applications for breast cancer diagnosis and treatment.
Jefferson Lab has long had this peripheral mission of reaching out and making its work relevant to the local community. So, we also have had students involved in educational outreach and K-12 teaching missions to excite the imagination of people who are not themselves inclined to be scientists. As we get closer to the greater variety of applications associated with the coming data science center, broader student and faculty outreach will create much larger communities of interest.
Another W&M strategic initiative involves training and practice in entrepreneurship and job creation. We hope to use internship opportunities for students at W&M to create virtuous cycles of economic uplift for businesses who want to access the untapped talent of our first-generation students from Virginia.
We want to absorb students from every disciplinary interest to have them engage with these advanced tools and learn to use them to maximal advantage for themselves and their employers. That’s our goal. We hope that no one anywhere will think that the JLab data center technology is not available to them. In fact, we want people to know that the less engaged you are with technology now, the more the coming AI tools can be helpful for you.
Q: The new facility at Jefferson Lab will be beneficial to all disciplines, but going back to nuclear physics — what should prospective students in this area expect when choosing William & Mary? What do they have to be aware of as the field evolves more and more rapidly?
A: These are questions that many of our recently hired nuclear physics faculty are answering. The key issues will be to continue to challenge the Standard Model and to look for subtle deviations or variations that might show we’ve been missing some fundamental physics along the way.
There also will be connections to be made between people who are interested in matter under extreme conditions of exceedingly high pressures or densities, or under exceedingly high or low temperatures. Folks interested in creating high magnetic fields will find kindred spirits with others who are interested in the creation of constrained matter, that is, matter in 1-dimension or 2-dimensions, where a small dislocation in curvature can cause electrons to behave in ways that are indistinguishable from very high magnetic fields. Much of this will come from connections to other national or international laboratories through the new center, which is to be the flagship for all data/computational matters in this realm. All of these scientific interests will translate into engineering improvements for the Electron-Ion Collider.
There is a lockstep in the progress that is self-evident but seems to escape some of the people who invest in research — the more you know, the more you will know. The more you do, the more you can do. So, these areas of fundamental science and engineering run step by step with nuclear physics: As you improve one thing, you automatically improve another. And that’s why the new hub will not simply be a repository for nuclear physics information. It will be a generator for a new approach for all information.
Antonella Di Marzio, Senior Research Writer