Specializing Cache Structures for High Performance and Energy Conservation in Embedded Systems

  • Authors:
  • Michael J. Geiger;Sally A. Mckee;Gary S. Tyson

  • Affiliations:
  • ECE and CIS Departments, University of Massachusetts Dartmouth North Dartmouth, MA 02747-2300,;Computer Systems Lab, Cornell University, Ithaca, NY 14853-3801,;Department of Computer Science, Florida State University, Tallahassee, FL 32306-4530,

  • Venue:
  • Transactions on High-Performance Embedded Architectures and Compilers I
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

Increasingly tight energy design goals require processor architects to rethink the organizational structure of microarchitectural resources. We examine a new multilateral cache organization, replacing a conventional data cache with a set of smaller region caches that significantly reduces energy consumption with little performance impact. This is achieved by tailoring the cache resources to the specific reference characteristics of each application. In applications with small heap footprints, we save about 85% of the total cache energy. In the remaining applications, we employ a small cache for frequently accessed heap data and a larger cache for low locality data, achieving an energy savings of 80%.