'High-Performance Computing Boot Camp' to Educate Faculty, Researchers at U.Va.

The University of Virginia will host a “High-Performance Computing Boot Camp” Aug. 7-16 to introduce faculty, graduate students and research professionals from across Grounds to the basics of high-performance parallel computing and the national cyber-infrastructure — and how they might be useful in their research. Held primarily on U.Va.’s Grounds, the workshop will conclude with a one-day trip to Virginia Tech’s Visualization Facilities in Blacksburg on Aug. 16. Computational science — the use of advanced computing capabilities to understand and solve complex problems — is a new wave of research methodology with a variety of applications and implications. “Computational science is one of the most important technical fields of the 21st century, because it provides a unique window through which researchers can investigate problems that are otherwise impossible to address — problems ranging from biochemical processes to weather patterns,” says James H. Aylor, dean of U.Va.’s School of Engineering and Applied Science. Leading the exploration of these implications will be camp instructors Andrew Grimshaw, U.Va. professor of computer science, and Nicholas Polys, postdoctoral associate at Virginia Tech’s Advanced Research Computing Center. Attendees will learn the basics of moving from sequential to parallel computing systems, and will come away with an understanding of the opportunities and challenges of data visualization tools and display technologies. Participants will also learn about the inner workings of supercomputers — how to optimize sequential applications, and how to locate and gain access to high-performance computing resources nationwide. This camp is one of the first steps the University is taking to answer a call issued in June 2005 by the federal President’s Information Technology Advisory Committee’s report, titled “Computational Science: Ensuring America’s Competitiveness.” The report spoke to the necessity of the comprehensive understanding and dissemination of this methodology as critical to scientific leadership, economic competitiveness and national security. In 2006, Aylor commissioned a Universitywide computational science initiative and task force. Led by Grimshaw; Mitch Rosen, chief technology officer at the Engineering School;, and John Hawley, chairman of the Department of Astronomy, the Task Force on Information Technology Infrastructure Supporting Research in Engineering and Science was charged with producing a set of recommendations that would ultimately improve the culture of computation at the University. In October 2006, the Engineering School received a two-year, $250,000 grant from the National Science Foundation to develop undergraduate and graduate courses in computational science. According to James Hilton, U.Va.’s chief information officer, who was instrumental in moving many of the task force’s suggestions forward and advancing U.Va.’s Computational Science Initiative, “This boot camp and the courses the NSF grant has made possible are the important milestones in leveraging experts and resources to bring computational science — in a very real way — to the University. The understanding of computational science is essential to enhancing U.Va.’s science and engineering capabilities and long-term technological leadership.”