Large, complex, multi-scale, multi-physics simulation codes, running on high performance computing (HPC) platforms, are essential to advancing science and engineering research in disciplines such as lattice field theory, astrophysics and cosmology, computational fluid dynamics/fluid structure interaction,and high energy density physics. Progress in computational science together with the adoption of high-level frameworks and modular development have produced widely used community simulation software specific to individual communities. These state-of-the-art codes have been under development and optimization for several years and currently simulate multi-scale, multi-physics phenomena with unprecedented fidelity on petascale platforms. Currently each of these codes have solvers with varied performance characteristics, but all face challenges because of changing hardware architecture. Efforts underway to cope with these challenges, are largely fragmented. While it is true that the scientific codes used in various domains differ significantly from one another, many solutions are likely to be conceptually similar, even if they differ in details. The goal of the proposed conceptualization project, Software Institute for Methodologies and Abstractions for Codes (SIMAC) is to find common abstractions and frameworks applicable across a broad range of applications through cooperation, coordination and interdisciplinary interactions among the participants. The core group of participating codes includes FLASH (astrophysics, cosmology, CFD, HEDP), Cactus (CFD, numerical relativity, and quantum relativity), the code suite used by the Lattice QCD community, and Enzo (cosmology).
The proposed collaborative research will produce benefit beyond the four simulation codes and collaborating institutions by exploring: a common software infrastructure applicable to a broad range of science and engineering application domains; an engagement model between computer science research and application development; a multidisciplinary immersion program for research, education and training of students, postdoctoral fellows and visitors on future platform architectures.
Academic and industrial scientists came together at University of Illinois NCSA for a workshop to discuss shared issues and challenges in developing and deploying software capable of running on supercomputers. This was workshop #2 for a series of 3 that were hosted. This specific workshop leveraged the attendees of NCSA's Private Sector Program, which were predominantly FORTUNE 100 companies in manufacturing, oil and gas, biomedical, and information technology. Topics of discussion included HPC platforms, optimization, programming models, algorithms, and performance. Particularly in the manufacturing sector, users of HPC sometimes fail to distiguish the reasons for poor software application performance, including a fundamental failure of determining whether the underlying science is adequate, or whether the algorithms do not scale, or perhaps whether the code architecture is old. Legacy commercial codes are prime suspects for poor architecture, and academic science codes are usually more modern and well architected for parallelism. These two communities came together to share success stories and possible avenues for collaboration. It was determined that academic/industry collaboration would be very valuable strategically, and a software innovation institute could be launched that would capture best practices from both communities, as well as capture multidisciplinary best practices that would lend promise to software enhancements for multiple sectors. Academic workshop leaders were from University of Illinois, Argonne National Lab, University of Chicago, Boston University, Michigan State University, Lawrence Berkeley Lab and Louisiana State University. Industry leaders attended from GE, Caterpillar, Boeing, Cray, Intel, Adaptive Computing, ExxonMobil, BP, Procter & Gamble, Dow Chemical, John Deere, Dell, 3M, NVIDIA, BHP Billiton, Rolls-Royce, Oracle, and several other IT and software development companies. NSF attendees included Irene Qualters and Dan Katz There were 112 total attendees. Agenda items included: - Future HPC platforms - NAMD optimizations on Blue Waters - Summary of workshop #1 - State of programming models and code transformations on heterogeneious platforms - Code performance: Is it the physics, algorithm, or Machine - Software Accountability for Sustainability, Support, IP - Performance Issues from an Industry Perspective/Research Perspective: Are they identical? - Metrics of Success in a Petascale World: What does at-scale mean? - A software development SWAT team perspective. Some surprising overlap was found in molecular dynamics codes, as well as in astrophysics and the physics used in manufacturing. Collaboration seemed to be valuable if these communities would come together and co-develop application software.
