Provisioning and Resource Allocation for Green Clouds

Abstract

The aim of Green Cloud Computing is to achieve a balance between the resource consumption and quality of service. In order to achieve this objective and to maintain the flexibility of the cloud, dynamic provisioning and allocation strategies are needed to regulate the internal settings of the cloud to address oscillatory peaks of workload. In this context, we propose strategies to optimize the use of the cloud resources without decreasing the availability. This work introduces two hybrid strategies based on a distributed system management model, describes the base strategies, operation principles, tests, and presents the results. We combine existing strategies to search their benefits. To test them, we extended CloudSim to simulate the organization model upon which we were based and to implement the strategies, using this improved version to validate our solution. Achieving a consumption reduction up to 87% comparing Standard Clouds with Green Clouds, and up to 52% comparing the proposed strategy with other Green Cloud Strategy.