Honolulu, HI, USA USA
June 24, 2012 to June 29, 2012
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/CLOUD.2012.16
Many cloud services nowadays are running on top of geographically distributed infrastructures for better reliability and performance. They need an effective way to direct the user requests to a suitable data center, depending on factors including performance, cost, etc. Previous work focused on efficiency and invariably considered the simple objective of maximizing aggregated utility. These approaches favor users closer to the infrastructure. In this paper, we argue that fairness should be considered to ensure users at disadvantageous locations also enjoy reasonable performance, and performance is balanced across the entire system. We adopt a general fairness criterion based on Nash bargaining solutions, and present a general optimization framework that models the realistic environment and practical constraints that a cloud faces. We develop an efficient distributed algorithm based on dual decomposition and the sub gradient method, and evaluate its effectiveness and practicality using real-world traffic traces and electricity prices.
Optimization, NIST, Resource management, Electricity, Load modeling, Games, Cloud computing, dual decomposition, datacenter selection, cloud
Hong Xu, Baochun Li, "A General and Practical Datacenter Selection Framework for Cloud Services", CLOUD, 2012, 2013 IEEE Sixth International Conference on Cloud Computing, 2013 IEEE Sixth International Conference on Cloud Computing 2012, pp. 9-16, doi:10.1109/CLOUD.2012.16