Nov. 7 — The Department of Defense (DOD) High Performance Computing Modernization Program (HPCMP) is pleased to announce the selection of the inaugural class of fiscal year 2014 Frontier projects. These projects explore science and technology and test and evaluation boundaries with computational approaches that would not have otherwise been possible. Awarded projects are expected to have a duration of three to five years each and be among the most computationally demanding projects the DOD will address.
“The Frontier projects push the boundaries of what the DOD’s science and engineering community can accomplish in support of the department,” said John West, director of the HPCMP.
“These projects represent an extraordinary leap ahead in the practice of advanced computing applied directly to the challenges facing the DOD,” commented Brad Comes, deputy director of the HPCMP. “The concept of Frontier projects is an evolution of our longstanding Challenge project program, but elevated to a new level. The awarded projects are planned to last up to five years and consume billions of hours of computer time. The HPCMP is providing support for the projects in the form of interdisciplinary subject matter expertise such as code optimization, scientific visualization, data management, workflow optimization and domain expertise. In addition, opportunities for the next generation workforce to engage with the projects are planned.”
The first award for fiscal year 2014 was made to Dr. Ryan Gosse of the Air Force Research Laboratory. Gosse’s team is pursuing the calculation of fully-coupled, high fidelity fluid-thermal-structural responses of vehicles in high mach number flight. “This project will be conducting first-of-its-kind calculations that only recently are theoretically tractable on the scale of systems available through the DOD HPCMP,” said Gosse. “The Air Force is currently studying options to build high-speed flight vehicles for a variety of mission needs, and fully-coupled high fidelity computational modeling is vital to this process as full-scale experimental testing is cost prohibitive with today’s physical testing technologies.”
The second award was made to Dr. Stephen M. de Bruyn Kops from the University of Massachusetts-Amherst, working on behalf of the Office of Naval Research. de Bruyn Kops and his team are performing simulations to aid in the study of the evolution of turbulence at high Reynolds numbers relevant to engineering problems in the ocean or atmosphere. “This proposed research will develop and deliver to DOD a set of predictive models to enable fast and accurate simulations of turbulence affected by gravity, as well as provide improved turbulence measurements at high Reynolds number, which are the ultimate verification of turbulence models,” said de Bruyn Kops. “Both components will be inputs to major defense efforts to develop next generation submarines and systems, fleet operation models, and advanced sensor systems and battlefield scale weather prediction models.”
This is the first year Frontier projects have been awarded by the HPCMP; additional projects will be awarded in subsequent years. The initial two projects were selected from a pool of nearly 60 proposals following an extensive and highly-competitive review process that considered scientific contribution, impact to the DOD, technical merit, and computational feasibility. The projects officially began at the start of the government’s fiscal year on Oct.1, 2013.
Source: Department of Defense