Software Engineering for Computational Science and Engineering on Supercomputers (SC16 BOF)

A Birds of a Feather session at SC16, on Thursday 17 November 2016

During the BOF

Please complete the evaluation form.  Your feedback contributes to the committee’s decision whether to have another BOF like this next year.

Jump to: Abstract |Agenda | Lightning Talks | Discussion Notes | Organizers | Suggested Citation


Software engineering (SWE) for computational science and engineering (CSE) is challenging, with more sophisticated, higher fidelity simulation of larger and more complex problems involving larger data volumes, more domains, and more researchers. Targeting high-end computers multiplies these challenges. We invest a great deal in creating these codes, but we rarely talk about that experience. Instead we focus on the results.

Our goal is to raise awareness of SWE for CSE on supercomputers as a major challenge and to begin the development of an international “community of practice” to continue these important discussions outside of annual workshops and other “traditional” venues.


Time Topic/Speaker
3 min Introduction and Goals
David E. Bernholdt, Oak Ridge National Laboratory
10 min Keynote Talk

Ganesh Gopalakrisnhan, University of Utah
Abstract: HPC software is characterized by its inclusion of domain mathematics, parallelism for scale, and longevity (when successful). Formal methods have an important role to play in helping avoid bugs, but the dosage depends on the scale of application (e.g. library functions or whole applications). We will briefly showcase two projects, one where a higher dosage was possible for tuning the precision of small floating-point routines, and another where large-scale data race checking of OpenMP programs relied more on good tool engineering.
27 min Lightning Talks (details below)
40 min General Discussion (notes below)
10 min Wrap-Up and Next Steps

Lightning Talks

Title/Speaker, Affiliation
1 Software Citation Principles
Daniel S. Katz, University of Illinois
2 Collaborating with academics to build software: some ways to fail
James Hetherington, University College London
3 Software Fellowship Programme (UK Software Sustainability Institute)
Aleksandra Pawlik, New Zealand e-Science Infrastructure
4 Is generic HPC Carpentry possible? Experiences from the community
Aleksandra Pawlik, New Zealand e-Science Infrastructure
5 Practical Software Sustainability @ The Netherlands eScience Center
Jason Maassen, Netherland e-Science Center
6 SC17: initiatives to improve inclusion in HPC
Toni Collis, EPCC
7 NSF Program Perspectives on Software Engineering in Science Software Projects
Rajiv Ramnath, National Science Foundation (US)
8 The Science Gateways Community Institute
Nancy Wilkins-Diehr, SDSC/UCSD
9 The ACME Climate Project Learning Initiative: A Cheatsheet
Mike Heroux, Sandia National Laboratories

Discussion Notes


David E. Bernholdt Oak Ridge National Laboratory
Jeffrey Carver University of Alabama
Neil Chue Hong University of Edinburgh
Mike Heroux Sandia National Laboratorie
Daniel S. Katz University of Illinois
James Lin Shanghai Jiao Tong University
Kengo Nakajima University of Tokyo

Suggested Citation

David E. Bernholdt, Jeffrey Carver, Neil Chue Hong, Mike Heroux, Daniel S. Katz, James Lin, and Kengo Nakajima, organizers, Birds of a Feather session on Software Engineering for Computational Science and Engineering on Supercomputers, in International Conference for High Performance Computing, Networking, Storage and Analysis (SC ’16), Salt Lake City, Utah, November 2015, URL:

SC16 program page