持続可能なデータセンター管理のためのカーボンアウェア・コンピューティング
Carbon-Aware Computing for Sustainable Data Centre Management (原題)
G. B. Sophia Shalini
🤖 gxceed AI 要約
日本語
データセンターの電力消費が2024年の415TWhから2030年に945TWhへ倍増する見通しの中、カーボンアウェア・コンピューティングの枠組みを45件の文献・業界報告・政策文書から整理した系統的レビュー。動的ワークロードスケジューリング、地理的マイグレーション、再エネ対応タスク配分、インテリジェント冷却、AI予測型エネルギー管理、循環経済の各戦略を比較し、クラウド設計者・政策担当者・AI開発者向けの実務指針を示す。
English
A systematic review of 45 studies, industry reports and policy documents (2019–2026) on carbon-aware computing for data centres, whose electricity demand is projected to nearly double from 415 TWh in 2024 to 945 TWh by 2030. It compares dynamic workload scheduling, geographic migration, renewable-aware allocation, intelligent cooling, AI-based predictive energy management and circular economy practices, offering practical guidance for cloud architects, policymakers and sustainability teams.
Unofficial AI-generated summary based on the public title and abstract. Not an official translation.
📝 gxceed 編集解説 — Why this matters
日本のGX文脈において
生成AI普及に伴うデータセンター電力需要の急増は、日本でもScope 2排出量と再エネ調達の課題として顕在化している。SSBJ開示や有報でのエネルギー関連開示を進める事業者にとって、カーボンアウェア運用は排出削減と調達戦略を結びつける実務的な選択肢となる。
In the global GX context
As generative AI drives data-centre electricity demand toward 945 TWh by 2030, carbon-aware computing links operational decisions to Scope 2 accounting and renewable procurement. It speaks to ISSB/CSRD disclosure of energy use and to transition planning for digital infrastructure, an area still under-served in global disclosure scholarship.
👥 読者別の含意
🔬研究者:Provides a structured taxonomy of carbon-aware computing strategies and identifies gaps for empirical validation of emissions-reduction claims.
🏢実務担当者:Offers a menu of operational levers—workload scheduling, migration, cooling, renewable-aware allocation—that data-centre and IT sustainability teams can assess against Scope 2 targets.
🏛政策担当者:Highlights the need for grid-carbon-intensity data, reporting standards and incentives that reward carbon-aware scheduling in digital infrastructure policy.
📄 Abstract（原文）
The rapid expansion of data centres for cloud computing and generative AI has resulted in unprecedented growth in global energy consumption, with electricity use expected to nearly double from 415 TWh in 2024 to 945 TWh in 2030. Carbon-aware computing has emerged as an important paradigm to reconcile digital growth with planetary sustainability. This systematic literature review explores carbon-aware computing frameworks for sustainable data centre management, considering 45 peer-reviewed studies, industry reports and policy documents published between 2019 and 2026. The review uses thematic synthesis to compare dominant strategies including dynamic workload scheduling, geographic workload migration, renewable-aware task allocation, intelligent cooling, AI-based predictive energy management and circular economy practices. The findings provide cloud architects, policymakers, AI developers and sustainability professionals with practical guidance for the transition to green computing.
🔗 Provenance — このレコードを発見したソース
- openalex https://doi.org/10.4018/979-8-2600-2025-8.ch004first seen 2026-10-23 05:23:22 · last seen 2026-11-06 04:58:49
🔔 こうした論文の新着を逃したくない方は キーワードアラート に登録（無料・3キーワードまで）。
gxceed は公開メタデータに基づく研究支援データセットです。要約・翻訳・解説は AI 支援で生成されています。 最終的な解釈・検証は利用者が原典資料に基づいて行うことを前提とします。