Optimal resource provisioning for scaling enterprise applications on the cloud

Abstract

Over the past years organizations have been moving their enterprise applications to the cloud to take advantage of cloud's utility computing and elasticity. However, in enterprise applications or workflows, generally, different components/tasks will have different scaling requirements and finding an ideal deployment configuration and having the application to scale up and down based on the incoming requests is a difficult task. This paper presents a novel resource provisioning policy that can find the most cost optimal setup of variety of instances of cloud that can fulfill incoming workload. All major factors involved in resource amount estimation such as processing power, periodic cost and configuration cost of each instance type and capacity of clouds are considered in the model. Additionally, the model takes lifetime of each running instance into account while trying to find the optimal setup. Benchmark experiments were conducted on Amazon cloud, using a real load trace and through two main control flow components of enterprise applications, AND and XOR. In these experiments, our model could find the most cost-optimal setup for each component/task of the application within reasonable time, making it plausible for auto-scaling any web/services based enterprise workflow/application on the cloud.