Managing energy, performance and cost in large scale heterogeneous datacenters using migrations
- Submitting institution
-
The University of Surrey
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 9001338_1
- Type
- D - Journal article
- DOI
-
10.1016/j.future.2018.10.044
- Title of journal
- Future Generation Computer Systems
- Article number
- -
- First page
- 529
- Volume
- 93
- Issue
- 0
- ISSN
- 0167-739X
- Open access status
- Compliant
- Month of publication
- -
- Year of publication
- 2018
- URL
-
-
- Supplementary information
-
-
- 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
-
-
- Research group(s)
-
-
- Citation count
- 8
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- The paper demonstrates realistic energy efficiency improvements for workload migration in large computer clusters. Improvements claimed in literature elsewhere are largely overstated due to several unrealistic assumptions regarding hardware and workload homogeneity, and cost-free mechanisms, using relatively small scale experiments; here, verification addresses data of 25m tasks across 12,500 machines - data from Google production systems - uniquely mapped to performance data collected for a multiplicity of heterogeneous cloud server hardware and workloads. The research, supporting workload and cost management, is being applied to continuous service provision under 5G multi-access edge server migrations for autonomous vehicles in an EPSRC project.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -