Abstract:
The quality of software is among the most underrated aspects of scientific discovery. To some extent this is because computational science is still relatively young. In the early days so many approximations were made in the model that one was only looking for very coarse grain insights. While even then it was not a good practice to not scrutinize the correctness of one’s software, it was much harder to define what correctness meant. As scientific understanding has grown, so have the tools and mechanisms that allow one to scrutinize assumptions made about the model and the methods used for solving the models. The application codes being prepared for the forthcoming platforms under the exascale computing project will have vast computational resources at their disposal. It behooves them to use the resources well. So the software is scrutinized not just for meeting the performance goals, but also be for the quality of results it produces. To be exposed to such requirements is an unequaled opportunity for the cohort of faculty and students participating in this program to get first hand experience of rigor being exercised in computational science. An early exposure to thinking about software design for quality will shape the thinking of the cohort in a good direction.
Short Biography:
Anshu Dubey is a Computational Scientist with deep experience in design, architecture and sustainability of multiphysics scientific software used on high performance computing platforms. Her contributions in the area of software engineering for research scientific software are widely known and respected. She has been the chief software architect for FLASH, a multiphysics multicomponent software that is used by several science domains including astrophysics, cosmology, solar physics, bio-mechanical systems, computational fluid dynamics and laser plasma experiments. She serves on the editorial boards of some of the leading journals in high performance computing. She has also served as the lead for Earth and Space Science Applications in the Exascale Computing Project. Anshu Dubey received her B. Tech from Indian Institute of Technology, New Delhi in 1985, and Ph.D. in computer science from Old Dominion University in 1993. She then joined the University of Chicago Astronomy & Astrophysics Department as a research associate and was promoted to research scientist in 1998. In 2001 she joined the ASC/Flash Center as lead application programmer, where she became computer science/applications group leader in 2003, and associate director in 2010. From 2013 to 2015 she was on the staff at Lawrence Berkeley National Laboratory, where she served as work lead and computer systems engineer in the Applied Numerical Algorithms Group. In 2015 she joined the Mathematics and Computer Science Division at Argonne as a computational scientist. She is also a Senior Scientist in the Computer Science Department at the University of Chicago.
Motivation:
I have led a GAG in SIAM-CSE in the past and enjoyed the experience of interacting with students with diverse backgrounds. One of the nicer outcomes was that one of the students wanted to apply for immigration to the United States, and he reached out to me for a letter of reference. It was gratifying to be able to make a positive difference in someone’s life. I would like to continue to be involved in such activities wherever possible.