0
RESEARCH PAPERS

The Exergy Cost of Information Processing: A Comparison of Computer-Based Technologies and Biological Systems

[+] Author and Article Information
V. P. Carey

Department of Mechanical Engineering, University of California, Berkeley, CA 94720-1740vcarey @ me.berkeley.edu

A. J. Shah

Department of Mechanical Engineering, University of California, Berkeley, CA 94720-1740amipshah@me.berkeley.edu

J. Electron. Packag 128(4), 346-352 (Dec 06, 2005) (7 pages) doi:10.1115/1.2351899 History: Received April 27, 2005; Revised December 06, 2005

Processing information (analysis, storing, retrieving, sharing) is the primary function of modern computer-based information systems. Systems of this type generally require an input flow of exergy (available energy) to function. Information processing systems now are evolving in two directions. One direction is toward bigger and more sophisticated systems. The other is toward systems that are more compact and portable. In both cases, the energy efficiency is becoming an increasingly important design issue. This paper summarizes an exploration of the exergy cost of processing information at the component and system levels in state-of-the-art information processing systems. The energy efficiency characteristics of computer-based information technologies are also compared to estimates of the energy efficiency of biological information processing in brains of mammals. Energy efficiencies of processors and systems are quantified in terms of the ratio of processing capacity to the exergy input rate. Available data suggest that for recent generations of processors, the ratio of processing capacity to exergy input rate has been increasing proportional to the square root of processor speed. Despite this increase, the energy efficiency of computer-based systems is currently substantially below the estimated efficiency of biological systems. Unless processor energy efficiencies are greatly increased, the development of information processing systems that match human brain performance will be hindered by the need for large power supplies and high-capacity heat rejection systems.

FIGURES IN THIS ARTICLE
<>
Copyright © 2006 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Figure 1

Performance of various microprocessor groups listed in Table 1

Grahic Jump Location
Figure 2

Typical (nonstartup) component exergy consumption rates for a desktop computer

Grahic Jump Location
Figure 3

Comparison of exergy consumption rates between different desktop systems

Grahic Jump Location
Figure 4

Performance of sample desktop systems

Grahic Jump Location
Figure 5

Estimated information processing performance of organisms versus metabolic exergy consumption

Grahic Jump Location
Figure 6

Estimated brain information processing performance of organisms versus metabolic exergy consumption of the brain

Grahic Jump Location
Figure 7

Estimated information processing efficiency variation with processor capacity

Grahic Jump Location
Figure 8

Projections of exergy consumption required for human brain performance

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In