Advanced mathematical methods for high-performance computational physics- an enabling CPU-GPU parallel hardware infrastructure

Abstract

This proposal seeks funding for the acquisition of a 4-node CPU-GPU (Central-Graphical Processing Unit) computer cluster, together with the necessary additional hardware and InfiniBand interconnect. Each one of the 4 nodes will contain 2 64-Core CPUs (AMD Milan 7713 2.0GHz) and 1TB of memory, as well as 8 GPUs cards (NVIDIA Tesla A100 PCIe v4 80GB). The computer will be placed in a server room which already contains the PIs existing 30-node cluster, and which has the necessary of power and cooling availability.

Document Details

Document Type
DoD Grant Award
Publication Date
Feb 29, 2024
Source ID
FA95502310180

Entities

People

  • Oscar Bruno

Organizations

  • Air Force Office of Scientific Research
  • California Institute of Technology
  • United States Air Force

Tags

Readers

  • Parallel and Distributed Computing.
  • Research Science/Academic Research