An autonomic approach for resource provisioning of cloud services

作者：Ghobaei Arani Mostafa^*; Jabbehdari Sam; Pourmina Mohammad Ali

来源：Cluster Computing, 2016, 19(3): 1017-1036.

DOI：10.1007/s10586-016-0574-9

• Cloud computing
• Cloud services
• Resource provisioning
• MAPE loop
• Learning automata

Recently, there has been a significant increase in the use of cloud-based services that are offered in software as a service (SaaS) models by SaaS providers, and irregular access of different users to these cloud services leads to fluctuations in the demand workload. It is difficult to determine the suitable amount of resources required to run cloud services in response to the varying workloads, and this may lead to undesirable states of over-provisioning and under-provisioning. In this paper, we address improvements to resource provisioning for cloud services by proposing an autonomic resource provisioning approach that is based on the concept of the control monitor-analyze-plan-execute (MAPE) loop, and we design a resource provisioning framework for cloud environments. The experimental results show that the proposed approach reduces the total cost by up to 35 %, the number of service level agreement (SLA) violations by up to 40 %, and increases the resource utilization by up to 25 % compared with the other approaches.

• 出版日期2016-9