Price modeling of laaS providers using multiple regression

Abstract

An alternative for users to reduce costs of acquire and maintain computational infrastructure to develop, implement and execute software applications is cloud computing. Cloud computing services are offered by providers and can be classified into three modalities: Platform-as-a-Service (PaaS), Software-as-a-Service (SaaS) and Infrastructure-as-a-Service (IaaS). In IaaS, the providers offer services divided into instances. With this, the user has a virtual machine at their disposal with the computational resources desired at a given cost. The main challenge faced by companies is to choose what is the best pricing plan (instance/provider) to supply their computational demand. Frequently, these companies need a large computational infrastructure to manage and improve their business processes and, due to the high cost of maintaining local infrastructure, they have begun to migrate applications to the cloud. This work aims to provide insights that can help companies in selection process of the best provider/instance to deploy and execute integrations solutions in the cloud. For this, a preliminary study to construction of a new proposal for price modeling of instances of virtual machines using linear regression is presented. In this approach, we consider the providers Amazon EC2, Google Compute Engine and Microsoft Windows Azure.