On the Limitations of Carbon-Aware Temporal and Spatial Workload Shifting in the Cloud
-
1.
University of Massachusetts Amherst
-
2.
Massachusetts Institute of Technology
Description
The aim of this repository is to provide the code to reproduce the results of the following work:
Sukprasert, Thanathorn and Souza, Abel and Bashir, Noman and Irwin, David and Shenoy, Prashant, "On the Limitations of Carbon-Aware Temporal and Spatial Workload Shifting in the Cloud", in the 19th European Conference on Computer Systems (EuroSys)
In this work, we conduct a detailed trace-driven analysis to understand the benefits and limitations of spatiotemporal workload scheduling for cloud workloads with different characteristics, e.g., job duration, deadlines, etc., based on hourly variations in energy's carbon-intensity over three years across 123 distinct regions, which encompass most major cloud sites. For more information, please refer to the paper.
Files
Files
(6.6 GB)
| Name | Size | Download all |
|---|---|---|
|
md5:9ccf295296496d5d27499442b1e33aec
|
6.6 GB | Download |
Additional details
Software
- Repository URL
- https://github.com/umassos/decarbonization-potential.git
- Programming language
- Python
- Development Status
- Active
References
- Sukprasert, Thanathorn and Souza, Abel and Bashir, Noman and Irwin, David and Shenoy, Prashant, "On the Limitations of Carbon-Aware Temporal and Spatial Workload Shifting in the Cloud", in the 19th European Conference on Computer Systems (EuroSys)