Graphics Processing Unit (GPU) Infrastructure for Massively Parallel Computing Research

Abstract

The objective of this DURIP proposal is to acquire graphics processing unit (GPU) computing infrastructure to support research projects involving massive parallelism at the University of Kansas (KU). We propose to acquire GPU infrastructure that together will form a single system for massively parallel computing research. The GPU cluster will be composed of 10 compute nodes with a total of 400 CPU cores and 20 NVIDIA GPGPUÕs. Our proposed infrastructure will leverage and extend existing KU infrastructure that is available to the researchers through the KU Advanced Computing Facility. The GPU infrastructure will support general-purpose graphics processing unit (GPGPU) research in five DOD mission-critical thematic areas: computational fluid dynamics, nanostructured materials, machine learning, unstructured meshing, and materials chemistry. The proposed infrastructure will enhance the quality of research and research-related education that is currently supported by the Department of Defense, is under review at the Department of Defense, and establishment of novel research capabilities in areas of research of interest to the Department of Defense. The proposed infrastructure will enable several research projects including: large eddy simulation; optical interactions in nanostructured materials; modal decomposition, reduced-order modeling, and phase topology of complex flows; machine learning for unmanned aircraft systems (UAS) and parallel planning; parallel moving meshes; parallel high-order meshes; simulations of solid-liquid interfaces, gas separations, and gas expanded liquids.

Document Details

Document Type
DoD Grant Award
Publication Date
Jul 09, 2020
Source ID
W911NF2010065

Entities

People

  • Suzanne M. Shontz

Organizations

  • Army Contracting Command
  • United States Army
  • University of Kansas

Tags

Readers

  • Computational Fluid Dynamics (CFD)
  • Parallel and Distributed Computing.
  • Research Science/Academic Research

Technology Areas

  • AI & ML
  • AI & ML - DoD AI Strategy
  • Autonomy
  • Microelectronics