Interference of billing and scheduling strategies for energy and cost savings in modern data centers
- Submitting institution
-
The University of Reading
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 83359
- Type
- D - Journal article
- DOI
-
10.1016/j.suscom.2019.04.003
- Title of journal
- Sustainable Computing: Informatics and Systems
- Article number
- -
- First page
- 49
- Volume
- 23
- Issue
- -
- ISSN
- 2210-5379
- Open access status
- Compliant
- Month of publication
- -
- Year of publication
- 2019
- URL
-
-
- Supplementary information
-
10.17632/8gk5ngsdst.1
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
3
- Research group(s)
-
7 - ACES
- Citation count
- 2
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- The conducted analysis shows that the state-of-the-art research of data centre power management is misguided as it only optimizes a fraction of the total cost of ownership. Understanding and optimizing energy efficiency for data centres has been a pressing research problem but most researchers omit the perspective of costs; this holistic analysis is a key contribution of the paper illustrating how research should be directed into areas that matter.
The article shows that the significant potential savings stem from simply shutting down idle nodes in batch operation, a strategy that is very rarely deployed in data centres.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -