Characterization of memory energy behavior

  • Authors:
  • H. S. Kim;M. Kandemir;N. Vijaykrishnan;M. J. Irwin

  • Affiliations:
  • Pennsylvania State Univ., University Park;Pennsylvania State Univ., University Park;Pennsylvania State Univ., University Park;Pennsylvania State Univ., University Park

  • Venue:
  • Workload characterization of emerging computer applications
  • Year:
  • 2001

Quantified Score

Hi-index 0.00

Visualization

Abstract

There is a growing consensus that energy consumption is becoming an important issue for a wide variety of systems. Two factors fuel these trends: (1) increasing use of embedded and portable computing devices in everyday life, and(2) very high power dissipation levels reached by current state-of-the-art processors. Consequently, recent years have witnessed a host of studies that address the problem of reducing the energy consumption. Unfortunately, most of the energy-oriented studies are from design automation and architecture domain, and only very recently soft-based approaches to the problem have appeared in research papers. In this chapter, we take an early step in evaluating the energy consumption of a selected set of benchmarks from three different application domains. Setting off with the observation that memory energy behavior and observe the variations in cache and memory energies under different cache configurations.