Skip to main content
v2026.11,772 entries · CC-BY 4.0
Dictionary termTrack EProposedv2026.2

Energy proportionality (computing)

The property of computing systems whose energy consumption scales linearly with workload, so that idle or lightly utilised systems draw correspondingly little power.

ByCASRAI Editorial Board
· Last updated 22 Aug 2026
Share this

Ask CASRAI · included with Regulatory Radar

Ask about Energy proportionality (computing)

Ask CASRAI answers research-administration questions and cites the passages behind every claim — and says so when the corpus does not cover something, instead of guessing. It comes with a Regulatory Radar subscription at $29 a month, alongside the daily digest of regulatory changes and the dashboard of what changed.

150 questions a day, on this site, over the API, or inside your own tools through the CASRAI MCP server.

Everything CASRAI publishes — this page, the dictionary, the guides and the news — stays free to read, with no account and no card.

Examples

Worked examples

  • Is an instance

    A cluster consolidates idle login nodes overnight to a single host, exploiting energy proportionality.

  • Is an instance

    A research group migrates from dedicated workstations to a shared compute service with higher average utilisation.

Counter-examples

Looks similar, but isn't

  • Not an instance

    A server that draws 200 W idle and 220 W under load is non-proportional.

  • Not an instance

    Always-on legacy hardware kept running just in case is the opposite.

Editorial commentary

Energy proportionality, a term introduced by Barroso and Holzle (IEEE Computer, 2007), describes a computing system whose power draw scales in proportion to its utilisation — an idle system should draw close to zero power, and a system at 50% load should draw roughly half the power it draws at 100%. Barroso and Holzle’s original observation was that most real servers fell well short of this ideal, commonly drawing 50-70% of peak power even near-idle, because components such as memory, disks and power supplies do not scale down the way CPUs increasingly do.

Why it matters for research computing

Most computing infrastructure spends much of its life below full utilisation — a dedicated workstation or a lightly used departmental server is often idle overnight and on weekends — so non-proportional hardware wastes energy precisely during the periods it is least needed. Modern CPUs with deep sleep and low-power states have improved substantially since 2007, but storage, memory and network equipment still lag behind, meaning proportionality remains an incomplete property of most real systems rather than a solved problem.

What it motivates in practice

The principle is the underlying justification for several common research-computing decisions: consolidating workloads from several lightly used dedicated machines onto fewer, more heavily utilised shared servers, rather than running many machines mostly idle; aggressive use of low-power and sleep states for genuinely idle hardware; and right-sizing cloud instance allocations to actual, measured usage rather than provisioning for a peak that is rarely reached. It is closely related to, but distinct from, carbon-aware computing, which is about scheduling relative to grid carbon intensity rather than hardware power-draw behaviour at a given load.

Recognising non-proportional hardware

A practical test: a server drawing, say, 200W idle and 220W under full load is close to non-proportional — nearly all of its energy cost is fixed regardless of whether it is doing useful work, which is the condition the principle is meant to identify and correct. This reasoning underpins procurement decisions covered under sustainable HPC.

References

  • Barroso and Holzle, “The Case for Energy-Proportional Computing,” IEEE Computer, 2007.

Also known as

Power proportionality · Energy-proportional computing

Machine-readable encodings

Use in your systems

JATS XML <role> element
xml
<role vocab="credit"
      vocab-identifier="https://casrai.org/dictionary/"
      vocab-term="Energy proportionality (computing)"
      vocab-term-identifier="https://casrai.org/dictionary/term/energy-proportionality-computing" />
Schema.org DefinedTerm (JSON-LD)
json
{
  "@context": "https://schema.org",
  "@type": "DefinedTerm",
  "@id": "https://casrai.org/dictionary/term/energy-proportionality-computing",
  "name": "Energy proportionality (computing)",
  "identifier": "https://casrai.org/dictionary/term/energy-proportionality-computing",
  "description": "The property of computing systems whose energy consumption scales linearly with workload, so that idle or lightly utilised systems draw correspondingly little power.",
  "inDefinedTermSet": "https://casrai.org/dictionary/domain/sustainable-research#set",
  "url": "https://casrai.org/dictionary/term/energy-proportionality-computing",
  "sameAs": [
    "Power proportionality",
    "Energy-proportional computing"
  ],
  "license": "https://creativecommons.org/licenses/by/4.0/",
  "publisher": {
    "@id": "https://casrai.org/#organization"
  },
  "author": {
    "@id": "https://casrai.org/#editorial-team"
  },
  "datePublished": "2026-05-21T02:23:02",
  "dateModified": "2026-08-22T16:25:15",
  "inLanguage": "en-GB",
  "isAccessibleForFree": true
}

Referenced across the research world

University of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logoUniversity of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logo
  • University of Cambridge logo
  • Columbia University logo
  • Crossref logo
  • University of Edinburgh logo
  • Harvard University logo
  • University of Oxford logo
  • Princeton University logo
  • Stanford School of Medicine logo
  • University College London logo
  • ORCID logo

View CASRAI adoption →

Regulatory Radar

Stop finding out after the fact

$29/month, cancel anytime. Daily digest updates from our analysis, a dashboard holding the same items, and a cited assistant for everything they raise.

  • Federal Register, Federal Register+, Grants.gov, Regulations.gov, NSF News, UKRI, plus CASRAI’s own published content.
  • 72,264 indexed passages, and every answer cites the ones it drew on.