BRI A Co-Design Approach for Advances in Software and Hardware

Abstract

Investigations were performed on energy savings for GAMESS and HPC Benchmarks on various Intel processor architectures and their analysis using the Empirical Model Decomposition (EMD) method. A novel energy-saving strategy, which takes into account the DVFS granularity of the different processor types while choosing a ceiling frequency (or power limiting level) has been proposed and studied. Under the auspices of the AFOSR BRI grant, we conducted multiple studies that explore the viability of multiple ARMv8 processor architectures in terms of performance and energy efficiency.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 03, 2019
Accession Number
AD1096473

Entities

People

  • Mark S. Gordon

Organizations

  • Iowa State University

Tags

Communities of Interest

  • Advanced Electronics
  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Air Force Research Laboratories
  • Algorithms
  • Applied Mathematics
  • Computational Chemistry
  • Computational Chemistry Methods
  • Computational Science
  • Computer Programming
  • Computer Simulations
  • Computers
  • Electronic Structure Theory
  • Energy Consumption
  • Field Programmable Gate Arrays
  • High Performance Computing
  • Instruction Set Architecture
  • Molecular Dynamics
  • Quantum Chemistry
  • Software Development

Fields of Study

  • Computer science
  • Engineering

Readers

  • Computational Modeling and Simulation
  • Energy Conservation and Renewable Energy Engineering.
  • Integrated Circuit Design and Technology.