16. Embracing the Memory and I/O Walls for Energy-Efficient Scientific Computing

  1. Albert Y. Zomaya and
  2. Young Choon Lee
  1. Chung-Hsing Hsu1 and
  2. Wu-Chun Feng2

Published Online: 30 JUL 2012

DOI: 10.1002/9781118342015.ch16

Energy-Efficient Distributed Computing Systems

Energy-Efficient Distributed Computing Systems

How to Cite

Hsu, C.-H. and Feng, W.-C. (2012) Embracing the Memory and I/O Walls for Energy-Efficient Scientific Computing, in Energy-Efficient Distributed Computing Systems (eds A. Y. Zomaya and Y. C. Lee), John Wiley & Sons, Inc., Hoboken, NJ, USA. doi: 10.1002/9781118342015.ch16

Editor Information

  1. Centre for Distributed and High Performance Computing, School of Information Technologies, University of Sydney, NSW, Australia

Author Information

  1. 1

    Oak Ridge National Laboratory, Oak Ridge, TN, USA

  2. 2

    Virginia Tech, Blacksburg, Virginia, USA

Publication History

  1. Published Online: 30 JUL 2012
  2. Published Print: 3 AUG 2012

ISBN Information

Print ISBN: 9780470908754

Online ISBN: 9781118342015

SEARCH

Keywords:

  • memory and I/O walls, for energy-efficient scientific computing;
  • memory/power wall, technical obstacles for realizing high performance;
  • model associating memory-access with program performance, needed;
  • effectiveness and portability balance, design challenge in DVFS algorithm

Summary

This chapter contains sections titled:

  • Introduction

  • Background and Related Work

  • β-Adaptation: A New DVFS Algorithm

  • Algorithm Effectiveness

  • Conclusions and Future Work

  • References