Sandia National Laboratories
Sandia National Laboratories was an early adopter of massive parallel computing, specifically for scientific computing/simulations of physical phenomena. Today Sandia researchers continue with mission-centric scientific computing, while also designing hardware, system software, and algorithms for future extreme-scale systems. There is increasing interest in data-centric computing including algorithms for graphs and analysis of high-speed streams. The Center for Computing Research and related centers mainly hire PhD mathematicians, computer scientists, and computational scientists. In this talk, I will discuss a sample of HPC and moderately parallel applications and platforms, the skills necessary for that research, and some opinions on desired vs expected skills for new team members. I will explain why many of the skills one learns in computing Kindergarten (undergrad or earlier) are still important in active HPC research.
Biosketch
Dr. Cynthia Phillips is a senior scientist at Sandia National Laboratories. She received a B.A. in applied mathematics from Harvard University and a PhD in computer science from MIT. In her (almost) 26 years at Sandia National Laboratories she has conducted research in combinatorial optimization, algorithm design and analysis, and parallel computation. Her work has spanned theoretical analysis, general solver development, and application prototypes. Her most recent focus areas include distributed graph algorithms, social network analysis, algorithm co-design for future high-performance architectures, and (parallel) streaming graph algorithms.
Cindy has received a number of awards including an R&D 100 award in 2006 as a member of team that designed a node allocator for massively parallel computers. In 2008 she was a finalist for the Franz Edelman award for operations research impact, for work with the EPA and a national labs team on sensor placement in municipal water networks. She has served as an officer and conference chair for organizations in math, computer science, and operations research.
Although she has never been a tenure-track university professor, she has a long-standing interest in education from elementary school through graduate school. She has a research professor position at the University of New Mexico, has worked with many students in research projects, served on thesis committees, and has recruited PhD candidates for Sandia. She has participated in outreach efforts at all levels through early-mid career advising. Since Sandia's future depends on an excellent work force, she has tried to follow recent efforts to improve education in parallel computing.