In this special feature written by Mike Bernhardt from The Exascale Report, we honor Dona Crawford, the first woman to grace the ranks of our Rock Stars of HPC.
I first met Dona Crawford at SC’95 in San Diego when she was the conference Deputy Program Chair and the HPC Challenge Co-chair. Two years later, Dona was one of the most visible leaders in the HPC community as the General Chair for SC’97 in San Jose.
I have worked with many top corporate and agency executives during my 23 years in the HPC community, and I have met very few community leaders with the spirit, enthusiasm, and love of life that we see in Dona Crawford. From her days as one of the original leaders of the Accelerated Strategic Computing Initiative (ASCI) program, a national effort dating back to the early 90s, to her current position as Associate Director for Computation at Lawrence Livermore National Laboratory (LLNL) where she is responsible for a staff of roughly 900, she has built a tremendous following of loyal employees and close friends. I have heard numerous colleagues refer to Dona as a true leader who inspires and motivates with vision and passion. She is admired by her employees and peers, respected by her colleagues, and loved by her friends.
It is indeed a great pleasure to acknowledge and introduce you to Dona Crawford – a true Rock Star of HPC.
INSIDEHPC: You have such a rich history in this community and have been involved in so many milestone activities – what would you call out as one or two of the high points of your career – some of the things of which you are most proud?
There are a few career milestones that come to mind.
Dona Crawford (left) and her sister Gail (right) having just finished running/walking Bay to Breakers
I was one of the original leaders of the Accelerated Strategic Computing Initiative (ASCI), a national effort dating back to the early 90s to provide—at that time—teraFLOPS of computing and the associated environment for nuclear weapons scientists to use computer simulations instead of conducting underground nuclear tests to certify the safety, security and reliability of the stockpile. ASCI (now known as ASC—Advanced Simulation and Computing) signified a paradigm shift in science from test-based to modeling- and simulation-based validation.
In the early days of ASCI, few computing experts believed the program would be able to take high-performance computing from 50 gigaFLOP/s to 100 teraFLOP/s in 10 years. Nobody had broken the one-teraflop barrier at the time, so it was quite a tall order. I would walk into a room with colleagues from other institutions and many would scoff at me. It was uncomfortable, but it turns out there’s something to be said for believing in a “wild” idea and fighting for its success.
ASCI is the result of a very large team of dedicated people from DOE Headquarters, industry, academia, and the labs. ASCI brought together the computing expertise of Livermore, Los Alamos, and Sandia national laboratories and established the framework for advancing computing to where it is today at the labs, with each of the national labs working in partnership with industry to pursue different hardware approaches and applications software. ASCI changed the way the world thinks about computing, and I’m proud of the small role I had in its inception. ASC today continues to push the boundaries of computational science, demanding ever more capability in computing hardware for predictive science.
As part of ASCI, I helped establish the Academic Strategic Alliances Program (ASAP). The idea was to have leading-edge universities work on large, complex, multi-disciplinary problems to validate our simulation-based approach. The Alliances were extremely successful at accelerating new developments in simulation science and high-performance technologies for computer modeling. This type of working relationship is good for the discipline and for the HPC community, and it continues today.
From the beginning, it was clear a critical component of the initiative would be making the supercomputing resources easy to access and use from remote locations. I was part of the team that created the DisCom2 (Distance and Distributed Computing and Communication) strategy, which blended two strategic thrusts. Distance Computing extended high-performance computing to remote sites, while Distributed Computing developed an enterprise-wide integrated supercomputing environment capable of supporting DOE’s science and engineering requirements. DisCom2 took advantage of the ongoing revolution in commodity- and cluster-based high-performance computing, as well as adopting and expanding on the open software approach to cluster computing. I was very interested in seeing DisCom2 come to fruition since I was located at the smaller of the two Sandia National Laboratories’ locations at the time and was leading the network research and development activities.
In 1993, I co-founded InfoTEST (née the National Information Infrastructure Testbed), which was a partnership between academia, private industry, and the national laboratories that was a precursor to the World Wide Web and Internet commerce. I was working at Sandia-Livermore and knew that to be efficient and productive, we had to have a way to access the big computers and computing resources at the other labs (Sandia/New Mexico, Livermore, and Los Alamos). For InfoTEST, I led a group that tied together distributed computing resources throughout the country and then demonstrated the feasibility and effectiveness of national-level access. This development resulted in performance data and practical experience that was critical to the establishment of the Internet. This was also one of the early efforts in network-based computing, which proved the viability of the concept for what would become known as grid computing.
I’ve also had several “firsts” as a woman in computing and management. For instance, I noticed I’m the first insideHPC female “rock star.” I was also the first mid-level and the first top-level technical female manager at Sandia. In the early 80s, I was the first technical female staff member to reduce my workweek to spend more time with my two babies at home. I worked four days a week, and even though I was still putting in 40 hours, it was considered part time. My managers were initially reticent — there was an underlying fear that all mothers would exercise this “reduced workweek” option if it were available and the workforce would therefore be reduced and part of it become less productive. Of course, none of those things happened. I only worked that schedule for four months, but I really treasured having an extra day at home with my children.
I think we’ve made progress in gender equality in computing. To be perfectly honest, I never felt there was a “glass ceiling” that I needed to push through. I just did what I was good at, worked hard, and was rewarded. But it does not go unnoticed that I am still the only woman in the room in many meetings.
INSIDEHPC: What are your thoughts on how we can attract the next generation of HPC professionals into the community – and provide them with the experience-based training that they will need to be successful?
I think it’s a combination of marketing and education. You have to put resources— time and money—into stoking the pipeline, and you need to find a way to communicate the exciting parts of what you’re doing in a way that connects with aspiring young scientists.
It’s never too early to start talking to kids and encouraging their curiosity about science and computing. There’s a great program I’m involved with called Expanding Your Horizons that encourages young women to consider careers in science, technology, engineering, and math. More of these organized efforts directed at young people are needed.
Lawrence Livermore organizes and funds an excellent summer scholar program and postdoctoral research program that is well known in academia. Our students and postdocs interact with world-renowned scientists on new areas of research and are given access to some of the most advanced computing facilities in the world. If you take the time to show the next generation a path—one that is exciting, meaningful, and has staying power—a good number of them will follow it.
INSIDEHPC: What motivates you? What is your passion?
Hands down, my passion is helping people. Luckily, in my job I get to do that in various ways, not the least of which is the fact that computing technology can transform the way we live and help improve our relationship with mother earth.
I’m passionate about sustainability. Clean air, clean water and low-carbon emitting, sustainable energy are goals of the highest order. The computing community has a tremendous capability at its disposal. We can design a model that reflects the entire earth system, not just its individual parts. And we can present the system in a manner that is understandable and even compelling to the general public. Because the changes to the earth are playing out over decades, it’s hard to comprehend and convey the need for individual and collective change today. We as a nation can’t implement the sort of changes necessary to achieve a sustainable world if we as citizens do not clearly understand the problem.
Science diplomacy is another topic I’m passionate about. I just returned from an eye-opening trip to Saudi Arabia on behalf of CRDF Global. CRDF’s objective is to advance global peace and prosperity through international scientific and technological collaboration. In my opinion, nothing but good can come from nurturing a spirit of science and technology cooperation, supporting opportunities to strengthen research and education in universities abroad, and providing critical benefits to the global community. When researchers interact on objective topics, each subconsciously learns to understand how the other feels about subjective topics. Understanding one another at different levels is what helps promote peace. The bonus is that we can help improve our global standard of living through science and technology.
INSIDEHPC: What “non-HPC” hobbies or activities do you have? If you ever really have ‘time off’ – how do you spend it?
There is nothing I’d rather be doing than spending time with my family. I have two very successful children who have always been my top priority as well as my greatest joy and source of pride. I also love to cook for and spend time with my extended family, plus I have many good, long-lasting friendships.
Additionally, I enjoy traveling and meeting people from other cultures. In those situations, I try to be respectful of their social mores.
Because I like people, I guess I have a bit of a reputation of trying to get people to “cut loose.” The way I do that is by cutting loose myself. I won’t reveal my secrets—you’ll have to ask others in the community about some of my antics.
INSIDEHPC: Approximately how many conferences do you attend each year? What would you say is your percentage of travel?
I’m selective about the conferences I attend, but I never miss the SC (Supercomputing) conference. I’ve been involved as an organizer in some way or another since 1991 and I was the general chair in 1997. I also try to attend the International Supercomputing Conference (ISC) and the Salishan Conference on High-Speed Computing.
Travel is a big part of my job. I travel about 25% of the time. I’ve been to Washington, D.C. on 11 separate trips so far in 2010. I serve on advisory committees for the National Research Council, the National Science Foundation, and the Council on Competitiveness, and I’m on the board of directors for CRDF Global. I also serve on a number of industrial advisory committees and academic or laboratory review committees. These opportunities all require a commitment to travel.
My current travel schedule is paltry compared to 20 years ago. During the early ASCI years, when I worked at Sandia-Livermore, I was traveling 48–50 weeks a year. I was leading Sandia’s effort to consolidate its Livermore computing operations with its other location in New Mexico, so I commuted by plane every week between Livermore and Albuquerque, with frequent side trips to Washington, D.C. thrown in. For 10 years, I lived mostly in hotels. I was on a first name basis with the hotel and car rental people.
INSIDEHPC: How do you keep up with what’s going on in the community and what do you use as your own “HPC Crystal Ball?”
Mark Seager is my crystal ball.
It’s easy for me to keep up with what’s going on in HPC because I have a staff of 900 absolutely brilliant people moving in many different directions, and I get all that information filtered back to me.
INSIDEHPC: What do you see as the most exciting possibility of what we can hope to accomplish over the next 5-10 years through the innovative applications of HPC?
I’m not sure it’ll happen in 10 years, but there will come a day when all the various tools and technology platforms available—our iPads, cell phones, supercomputers, televisions—will merge into one big knowledge ecosystem. Technology will change our existence in ways that we can’t foresee today. That is exciting. With technology comes knowledge, knowledge breaks down fear, and fear is what causes trouble in the world. I think the most exciting possibility is that technology will help humans become more unified.
INSIDEHPC: What are your thoughts on HPC’s ability to address what many are referring to as “the missing middle” which I loosely interpret as a broad spectrum of small and mid-size businesses. (Is HPC starting to reach a larger audience of people who previously did not have access to it?)
The first thing we have to do is understand the problems that small and mid-size businesses are interested in solving, and then figure out practical ways HPC can address their needs. The next step is forming partnerships to help overcome some of the barriers: make the infrastructure affordable and the codes easy to use. With our history of building out the technology focused on solving specific problems, we can and should provide a bridge to companies looking to do the same. The national labs will always need to push the tip of the pyramid in HPC, but if we don’t help build out the base, there is a danger the tip will topple . Many of the labs, Livermore among them, are working with a wider variety of industrial partners, one at a time, to do just that.
INSIDEHPC: What do you see as the single biggest challenge we face over the next 5-10 years?
Power. We could easily go about making the next big supercomputer by stringing a bunch of components together, but if the resulting machine requires 100MW of power to operate, that’s just not a realistic option. We have to innovate machines that require substantially less power.
INSIDEHPC: Any closing thoughts you would like to share with the HPC community?
From left, Karli, Dona's son Josh, Dona, Dona's daughter Julia, and Mike
It has never just been about creating the computer with the most superlatives attached to it; it’s the discoveries the machines make possible. Supercomputers have become the backbone of science and technology, and the simulations performed on them will enable virtually every scientific field for decades to come.
For instance, climate modelers have said they need exascale computing capabilities to achieve high-resolution coupled earth system models at 1-km resolution. Having more knowledge about climate change and its effects earlier by even a few years may well be worth a billion dollars.
China’s new supercomputer recently took the world performance lead, and the country’s government and scientists should be applauded for this remarkable achievement. Because I work at a national security laboratory, I tend to think about future challenges in terms of national security. High performance computing and simulation are essential both for national security and for industrial competitiveness in the world economy. China’s major investments in HPC show that they recognize this truth and are willing and able to focus money, energy, and creativity in this direction. There are other similarly focused efforts in Russia, Europe, and Japan.
Technology is at an inflection point where the laws of physics are dictating that we do things differently. The underlying technologies exploited this past decade to build ever-faster supercomputers are facing the end of an era of “easy” gains. Technology will change and everything that technology touches will change.
If we cede our leadership on the hardware side, it’s very likely we’ll also eventually cede our leadership in software, components, and other critical technologies that support cost-effective and powerful server and PC markets all the way down to the cell phone. There is no doubt that these technologies provide distinct advantages to the sponsoring nation.
As a nation, we’ve given a lot away; let’s at least keep our innovation. If we’re going to continue to use HPC as an economic engine for competitiveness in the global marketplace, we need focused and consistent investments in advanced computing technology.
The old timers in the community of course know Steve Wallach. As co-founder of Convex Computer Corporation, he was well respected throughout the computational science and the investment communities. His technical leadership was chronicled in Tracy Kidder’s Pulitzer Prize winning book, “The Soul of a New Machine.” Convex was eventually acquired by Hewlett-Packard, and Steve took on the role as Chief Technology Officer of HP’s Enterprise Systems Group.
Steve Wallach: Well, the march toward exascale computing is upon us. The consensus is that getting to exascale will NOT be as straightforward as getting to petascale. I have some thoughts on this and will be presenting those thoughts at several SC10 panels. Clearly exascale will be discussed and debated.
Today Bill Kramer is the deputy project director and co-principal investigator for the Blue Waters project at the National Center for Supercomputing Applications (NCSA), at the University of Illinois in Urbana-Champaign. This is ground zero for the first sustained PFLOPS (10+ PFLOPS peak) supercomputing center dedicated to diverse science and engineering; but it’s not really about the computer. Over the past several years Bill and his team have been focused on building the facility and designing a system that, when finally turned on next year, will probably be the largest system for open science in the world. But if you’ve been following what the Blue Waters team has been doing you’ll see that they have taken a radically different approach to the launch of this capability into the community.
This is the perfect place for Bill Kramer. In talking with Kramer about his accomplishments, it is clear that he is one of those people who have driven their career paths with a guided purpose. As he describes it, the common thread across all of the places he’s been in his career is that they were all setting the pace for HPC at the time.
Throughout all of these very challenging assignments, Kramer has remained dedicated to volunteer service. “These are very symbiotic commitments,” he says. “Certainly the organizations benefit, and I enjoy giving back to the community. But volunteer assignments are a great way to refresh my point of view and to develop new skills that, sometimes, end up helping out professionally.” Kramer says that a lot of what he has learned about managing people has come from experience in volunteer organizations. Over the years he has served in SCUBA organizations and volunteered in schools and community theaters. He also helped start the tutorials effort and graphics special interest group of the Digital User’s Group, and has been active in SIGGRAPH. But people are probably most familiar with his service to the SC conference series, which included a year as General Chair of the Conference in 2005 when he hosted Microsoft Chairman Bill Gates on the stage in Seattle.
