(DURIP) OPTIMIZATION, CONTROL, NETWORKS AND LEARNING FROM DATA
Abstract
The goal of this project is to procure a GPU based cluster at George Mason University (GMU), which will help design software that efficiently allow development and application of solvers for the simulation, control, and optimization of processes involving complex multiphysics relevant to DoD. These types of simulations and control problems require large High Performance Computing (HPC) resources. Due to large body of institutions in the Washington DC area, there is a tremendous need and opportunity to create such a resource. The proposal identifies a large class of research problems, relevant to DoD, which cannot be handled using the existing or available hardware resources at GMU and the partner institutions. For instance- shape optimization, flow control (e.g. air-flow in hospitals, meeting rooms, or other DoD buildings where diseases can spread), replacing chemical reaction packages, equation of state calculators or other CPU-intensive ‘physics modules’ commonly encountered in legacy production codes by deep neural nets, control and optimization problems constrained by dynamical systems with uncertainty. Notice that many of the existing DoD legacy codes cannot take advantage of modern GPU based architectures and modern mathematical tools such as adjoint based optimization. Moreover, the current concerns about security at DoD machines have made it almost impossible for our students to access them. Finally, we have noticed that optimal coding and algorithm design is no longer taught in STEM programs. This implies a clear need for the machine sought under this proposal.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Mar 07, 2023
- Source ID
- FA95502110181
Entities
People
- Harbir Antil
Organizations
- Air Force Office of Scientific Research
- George Mason University
- United States Air Force