Sensitivity Analysis of Carbon-Aware Scheduling in Cloud Computing: A Simulation-Based Study

Data centres can trim their emissions by running flexible work while the grid is cleaner, but how large that trim appears depends heavily on the modelling choices behind the number. This study examines that dependence in a deliberately simple setting: ten synthetic jobs, 324 actual half-hourly Great Britain carbon-intensity readings across seven days (24–30 August 2026), a fixed 50 W per CPU core, and a scheduler given perfect foresight of the intensity that follows. With zero waiting allowed, estimated savings are nothing; grant up to 180 minutes and they reach about 4%. Those figures ignore each job's own deadline and the limited pool of cores. Introducing both with a mixed-integer formulation reduces the estimate to approximately 2.6%. Every value here is a deterministic best-case simulation rather than a measurement from a live data centre, so the short record, invented jobs, rough power assumption, and look-ahead signal all restrict how far the conclusions travel.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-18
DOI
https://doi.org/10.5281/zenodo.24346442
Primary Topic
Cloud Computing and Resource Management
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Sensitivity Analysis of Carbon-Aware Scheduling in Cloud Computing: A Simulation-Based Study

Manya Rathore, Jyoti Raut, Pranita Manjare, Aditi Kandari et al.
1 links
Zenodo (CERN European Organization for Nuclear Research)
Cloud Computing and Resource Management
article

Sensitivity Analysis of Carbon-Aware Scheduling in Cloud Computing: A Simulation-Based Study

Manya Rathore, Jyoti Raut, Pranita Manjare, Aditi Kandari, Sakshi Jamdade
article en

Abstract

Data centres can trim their emissions by running flexible work while the grid is cleaner, but how large that trim appears depends heavily on the modelling choices behind the number. This study examines that dependence in a deliberately simple setting: ten synthetic jobs, 324 actual half-hourly Great Britain carbon-intensity readings across seven days (24–30 August 2026), a fixed 50 W per CPU core, and a scheduler given perfect foresight of the intensity that follows. With zero waiting allowed, estimated savings are nothing; grant up to 180 minutes and they reach about 4%. Those figures ignore each job's own deadline and the limited pool of cores. Introducing both with a mixed-integer formulation reduces the estimate to approximately 2.6%. Every value here is a deterministic best-case simulation rather than a measurement from a live data centre, so the short record, invented jobs, rough power assumption, and look-ahead signal all restrict how far the conclusions travel.

Zenodo (CERN European Organization for Nuclear Research)
Climate action
Cloud Computing and Resource Management
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon