Features:

DAN REED: A PERSONAL PERSPECTIVE ON A CAREER IN HPC
by Alan Beck, Editor-in-Chief

After 20 years at the University of Illinois at Urbana-Champaign and four years as the director of the National Center for Supercomputing Applications (NCSA), Daniel A. Reed has accepted an appointment as founding director of a new interdisciplinary computing institute based at the University of North Carolina at Chapel Hill. Duke and North Carolina State universities are partnering with UNC on the institute. His start date is January 2004. HPCwire recently spoke with Dan Reed about his decision and its implications.

HPCwire: You've spent the last four years at NCSA and the last 20 years at the University of Illinois at Urbana-Champaign. How hard was the decision to leave after so many years and what were some of the factors in making that decision?

REED: It was certainly a difficult decision; one that took much time and serious thought. In the end, it was a very personal choice. During the past year, my father died of cancer and emphysema, after a long and valiant fight. Watching a parent cope with a debilitating and ultimately, fatal illness leads to serious reflection on one’s own life - accomplishments, goals, priorities and futures. Over the past ten years, I have regularly deflected queries about positions elsewhere. However, this time I thought about the fact that I had been at Illinois 20 years and that now was the time to launch a new phase in my life and my career.

HPCwire: What are the accomplishments you are most proud of and for which you will want to be remembered in NCSA and UIUC history?

REED: I am extremely proud of the role I played in securing new buildings for both NCSA and the University of Illinois computer science department. The Siebel Center for Computer Science, funded in large part by UI alumnus Thomas Siebel, is the focal point of a new campus technology quadrangle. It will give the department the chance to grow, in both size and scope. The new NCSA building will be the other star attraction of the technology quadrangle. It will bring the staff together in one location for the first time since NCSA opened its doors in 1986.

In addition, we have seen amazing accomplishments at NCSA during the past four years. The TeraGrid will begin operation in January as the largest open computing resource in the world. We have launched new projects with long-term impact on the nation’s security and scientific enterprise. Within NCSA, the new National Center for Advanced Secure Systems Research (NCASSR) is examining data and pattern discovery techniques, software defined radio, and network security techniques that will secure our critical information infrastructure.

Our collaboration with the National Archives and Records Administration (NARA) is developing tools for managing the deluge of digital data now engulfing us. We are also creating new cybercommunities via the National Science Foundation's NEESgrid (National Earthquake Engineering Simulation grid) project and the LEAD (Linked Environments for Atmospheric Discovery) project.

We have also made great progress in expanding the high-performance computing community through the NSF's Advanced Networking with Minority Serving Institutions (AN-MSI) program and through the new HASTAC effort (Humanities, Arts, Science, and Technology Advanced Collaboratory). The latter seeks to connect the arts and humanities with high-performance computing and networking, enabling new, collaborative modes of discovery.

NCSA has also had great success with outreach, thanks to scientific visualizations that have been incorporated into sky shows at the Hayden Planetarium in New York City, and used in TV documentaries for NOVA and the Discovery Channel.

HPCwire: When the leader of a major center like NCSA decides to move on, people naturally ask questions about that center's future. How does the future look for NCSA, its programs and its people?

REED: NCSA is stronger than any one director - no matter how visionary or charismatic. The true strength of any organization is its people, and NCSA is no exception. During the past four years I have been privileged to work with an incredible mix of talented, creative individuals who make the whole greater than the sum of its impressive parts. I charted some new directions, but the NCSA staff are the ones who make the center an incredible place, a place where the future is created every day.

On a more practical note, the funding picture for NCSA is as robust and stable as it has ever been at any time in the center's history, and opportunities abound for new projects in cybersecurity, data management and analysis, nanotechnology and the merging of the arts and technology.

In addition, I am leaving NCSA in the most capable hands of Interim Director Rob Pennington and Executive Director Danny Powell. Rob is one of the key players who made the TeraGrid vision a reality, truly a terascale task. Danny is the solid rock of NCSA - a reasoned voice of management. Together, they will guide the NCSA ship into an exciting future.

HPCwire: Even though you are leaving NCSA, you have set the direction for the center and formulated its vision for the last four years, and the effects of that leadership will be felt for years to come. Do you have any thoughts on where NCSA will be in five years, or even 10 years?

REED: Yogi Berra once said, “It's tough to make predictions, especially about the future.” For NCSA, though, predictions are easy - NCSA will be a leader in defining and implementing cyberinfrastructure and in broadening the base of scientists and engineering. High energy physics and cosmology, two seemly disparate fields, are now intertwined, as we ask deep questions about the origin of space and time. Biology and medicine are poised to answer questions older than humanity - the origin of life and its processes. NCSA will be the enterprise that works in collaboration to build the infrastructure that will answer these and other compelling questions.

HPCwire: What can you say about the new institute in North Carolina that you will lead? When will it actually be open for business? What will be its primary focus?

REED: In North Carolina, I will be the founding director of the Institute for Renaissance Computing, a venture jointly supported by the University of North Carolina at Chapel Hill, North Carolina State University and Duke University. The Institute will explore the interactions of computing with the arts and humanities, science and engineering, and the fabric of our daily lives. It will also partner with business leaders to enhance the competitiveness of North Carolina industries. The Institute will begin operation this spring, as faculty, staff and students are recruited.

Because many recent biological discoveries are computer-aided, one of my interests is marshaling computing talent to enrich computing collaborations with biomedical research. The biological revolution has just begun, and I believe bioinformatics will have an enormous impact on health and medical care.

Beyond biology, we want to unlock computing’s true power to enrich and drive discovery across the entire range of human activities. The Research Triangle campuses offer enormous potential for adapting technology to serve the arts and humanities, to catalyze scientific discovery, to shape public policy and to enrich the human experience via the novel application of computing and collaboration technology.

See http://www.unc.edu/news/newsserv/archives/dec03/reed121703.html for the announcement of the Institute.

HPCwire: NCSA has always stressed collaboration with other research institutions. Will the new institute also promote cross-disciplinary and multi-institutional work?

REED: Yes. The Institute is a broad-based activity that will build local, state, national and international collaborations. Borrowing a great idea from Donna Cox, head of NCSA's Experimental Technologies division, it will be based on the notion of Renaissance teams. A “Renaissance team” approach will bring scientists, engineers, artists and institute staff together to explore interdisciplinary approaches to scholarship, discovery and education.

HPCwire: Any other thoughts you would like to share with your friends and colleagues as you make this important life and career transition?

REED: Let me close with a reprise of what I said to the NCSA staff when I announced the transition. I quoted the famed Chicago architect and city planner Daniel Burnham, who said, “Make no little plans. They have no power to fire men's spirits.”

We are privileged to help turn dreams into plans and plans into reality. We must dream big - the true impact of computing on humanity has yet to be realized, and it is limited only by the size of our dreams.