Cloud elasticity - the ability to use as much resources as needed at any given time - and low cost - a user pays only for the resources it consumes - represent solid incentives for many organizations to migrate some of their computational activities to a public cloud. As the interest in cloud computing grows, so does the size of the cloud computing centers and their energy footprint. The realization that power consumption of cloud computing centers is significant and it is expected to increase substantially in the future motivates our interest in scheduling and scaling algorithms which minimize power consumption. We propose energy-aware application scaling and resource management algorithms. Though targeting primarily the Infrastructure as a Service (IaaS), the system models and the algorithms we propose can be applied to the other cloud delivery models and to private clouds.
View on arXiv