CloudOpt: multi-goal optimization of application deployments across a cloud

  • Authors:
  • Jim Zw Li;Murray Woodside;John Chinneck;Marin Litoiu

  • Affiliations:
  • Carleton University, Ottawa, Canada;Carleton University, Ottawa, Canada;Carleton University, Ottawa, Canada;York University, Toronto, Canada

  • Venue:
  • Proceedings of the 7th International Conference on Network and Services Management
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

CloudOpt is a comprehensive approach to find optimal deployments for large service centers and clouds. It uses a combination of bin-packing, mixed integer programming and performance models to make decisions affecting diverse and strongly interacting goals, including satisfaction of different service level agreements for many different applications (involving response time, workload, throughput), host memory requirements and availability, license costs and power consumption. It is scalable and extendable to new objectives.