aceModerate evidence
Published recordClimateGood Space
More
Towards Greener Data Centers: Integrated Optimization of Cooling and Resource Usage via Machine Learning
- Truvace record
- TRV-2026-2626raw data (JSON)
- Version
- v1Published
- First published
- Evidence type
- Benchmark or lab evaluation
- Sources
- 1 sourcePeer-reviewed
- sha256
- d01309f5b55b…check it
Data centers form the backbone of the global digital economy, yet their exponential growth has led to significant energy consumption, with cooling systems alone accounting for 30 to 50% of total energy usage.This paper proposes an integrated framework that combines machine learning based workload prediction with dynamic cooling control to achieve holistic energy optimization.The system employs a multi-layered architecture comprising real-time sensor telemetry, predictive analytics (Random For-est and Reinforceme…
The gain
In a simulation of a data center, a machine learning controller cut cooling energy by 28.1% on average and lowered PUE from 1.58 to 1.42 compared with a fixed-setpoint baseline.
What this doesn’t fix
The evaluation is simulation-based; real-world sensor noise, actuator delays and hardware differences are not fully captured, and the reinforcement learning agent uses a simplified state space.
Sources
1 on file- International Journal of Computer Applicationsdoi:10.5120/ijca2e7b62930f25
Cite this record
v1Truvace Impact Record TRV-2026-2626, v1: “Towards Greener Data Centers: Integrated Optimization of Cooling and Resource Usage via Machine Learning.” Truvace, 2026-11-12. /record/TRV-2026-2626 (accessed on the day you copy it). sha256 d01309f5b55b8761…
Change history
1 versionEvery change made to this record since it was first published, and why. Nothing has changed since it went up.
Certified into the record
Check this record hasn’t been altered
How to check it yourself, without trusting this page
Download any version from /api/record/TRV-2026-2626 and hash its canonical field yourself, for example with shasum -a 256. You should get its content_hash. Each version’s text also ends with prev:and the hash of the version before it (version 1 points to 64 zeros). If anyone had changed a single character of any version after it was saved, those hashes wouldn’t line up.