Inter-continental Data Centre Power Load Balancing for Renewable Energy Maximisation
The ever increasing popularity of Cloud and similar services pushes the demand for data centres, which have a high power consumption. In an attempt to increase the sustainability of the power generation, data centres have been fed by microgrids which include renewable generation—so-called ‘green dat...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-05-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/11/10/1564 |
| _version_ | 1797500317347086336 |
|---|---|
| author | Rasoul Rahmani Irene Moser Antonio L. Cricenti |
| author_facet | Rasoul Rahmani Irene Moser Antonio L. Cricenti |
| author_sort | Rasoul Rahmani |
| collection | DOAJ |
| description | The ever increasing popularity of Cloud and similar services pushes the demand for data centres, which have a high power consumption. In an attempt to increase the sustainability of the power generation, data centres have been fed by microgrids which include renewable generation—so-called ‘green data centres’. However, the peak load of data centres often does not coincide with solar generation, because demand mostly peaks in the evening. Shifting power to data centres incurs transmission losses; shifting the data transmission has no such drawback. We demonstrate the effectivity of computational load shifting between data centres located in different time zones using a case study that balances demands between three data centres on three continents. This study contributes a method that exploits the opportunities provided by the varied timing of peak solar generation across the globe, transferring computation load to data centres that have sufficient renewable energy whenever possible. Our study shows that balancing computation loads between three green data centres on three continents can improve the use of renewables by up to 22%. Assuming the grid energy does not include renewables, this amounts to a 13% reduction in CO<sub>2</sub> emissions. |
| first_indexed | 2024-03-10T03:00:10Z |
| format | Article |
| id | doaj.art-9516cbc85706408f9de764fbda5c1f1f |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-10T03:00:10Z |
| publishDate | 2022-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-9516cbc85706408f9de764fbda5c1f1f2023-11-23T10:47:02ZengMDPI AGElectronics2079-92922022-05-011110156410.3390/electronics11101564Inter-continental Data Centre Power Load Balancing for Renewable Energy MaximisationRasoul Rahmani0Irene Moser1Antonio L. Cricenti2Department of Computer Science and Software Engineering, Swinburne University of Technology, Melbourne, VIC 3122, AustraliaDepartment of Computer Science and Software Engineering, Swinburne University of Technology, Melbourne, VIC 3122, AustraliaDepartment of Computer Science and Software Engineering, Swinburne University of Technology, Melbourne, VIC 3122, AustraliaThe ever increasing popularity of Cloud and similar services pushes the demand for data centres, which have a high power consumption. In an attempt to increase the sustainability of the power generation, data centres have been fed by microgrids which include renewable generation—so-called ‘green data centres’. However, the peak load of data centres often does not coincide with solar generation, because demand mostly peaks in the evening. Shifting power to data centres incurs transmission losses; shifting the data transmission has no such drawback. We demonstrate the effectivity of computational load shifting between data centres located in different time zones using a case study that balances demands between three data centres on three continents. This study contributes a method that exploits the opportunities provided by the varied timing of peak solar generation across the globe, transferring computation load to data centres that have sufficient renewable energy whenever possible. Our study shows that balancing computation loads between three green data centres on three continents can improve the use of renewables by up to 22%. Assuming the grid energy does not include renewables, this amounts to a 13% reduction in CO<sub>2</sub> emissions.https://www.mdpi.com/2079-9292/11/10/1564data centre load balancinggreen data centremulti-agent systems |
| spellingShingle | Rasoul Rahmani Irene Moser Antonio L. Cricenti Inter-continental Data Centre Power Load Balancing for Renewable Energy Maximisation Electronics data centre load balancing green data centre multi-agent systems |
| title | Inter-continental Data Centre Power Load Balancing for Renewable Energy Maximisation |
| title_full | Inter-continental Data Centre Power Load Balancing for Renewable Energy Maximisation |
| title_fullStr | Inter-continental Data Centre Power Load Balancing for Renewable Energy Maximisation |
| title_full_unstemmed | Inter-continental Data Centre Power Load Balancing for Renewable Energy Maximisation |
| title_short | Inter-continental Data Centre Power Load Balancing for Renewable Energy Maximisation |
| title_sort | inter continental data centre power load balancing for renewable energy maximisation |
| topic | data centre load balancing green data centre multi-agent systems |
| url | https://www.mdpi.com/2079-9292/11/10/1564 |
| work_keys_str_mv | AT rasoulrahmani intercontinentaldatacentrepowerloadbalancingforrenewableenergymaximisation AT irenemoser intercontinentaldatacentrepowerloadbalancingforrenewableenergymaximisation AT antoniolcricenti intercontinentaldatacentrepowerloadbalancingforrenewableenergymaximisation |