Green-aware workload scheduling in geographically distributed data centers
Renewable (or green) energy, such as solar or wind, has at least partially powered data centers to reduce the environmental impact of traditional energy sources (brown energy with high carbon footprint). In this paper, we propose a holistic workload scheduling algorithm to minimize the brown energy...
Saved in:
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Conference or Workshop Item |
| Language: | English |
| Published: |
2013
|
| Online Access: | https://hdl.handle.net/10356/97268 http://hdl.handle.net/10220/13075 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | Renewable (or green) energy, such as solar or wind, has at least partially powered data centers to reduce the environmental impact of traditional energy sources (brown energy with high carbon footprint). In this paper, we propose a holistic workload scheduling algorithm to minimize the brown energy consumption across multiple geographically distributed data centers with renewable energy sources. While green energy supply for a single data center is intermittent due to daily/seasonal effects, our workload scheduling algorithm is aware of different amounts of green energy supply and dynamically schedules the workload across data centers. The scheduling decision adapts to workload and data center cooling dynamics. Our experiments with real workload traces demonstrate that our scheduling algorithm greatly reduces brown energy consumption by up to 40% in comparison with other scheduling policies. |
|---|