VBalance: A Selection Policy of Virtual Machines for Load Balancing in Cloud Computing

摘要

Cloud computing is a new computing paradigm, in which the data processing and data storage are offered to customers similar to services accessed over the Internet. These services are represented by Virtual Machines (VMs) which are hosted by real servers located in large data centers. The Virtual Machines share the physical resources offered by servers. The growing demand for solutions related to Cloud has resulted in greater investments in electrical power for service providers. Today, this financial cost represents one of the greatest expenditures incurred by maintaining such data centers. Therefore, solutions have been proposed to minimize power expenditure while maintaining maximum care with requirements demanded by customers. This work presents a selection policy project utilized for the balance of workloads among servers in a data center. The proposed policy will address power issues during its selection process. The VBalance policy will select for migration, the VM that contributes most to the power consumption of the server by using the Multiple Linear Regression (MLR) model. Furthermore, a comparative analysis is made of the proposed solution with other existing models, in order to evaluate the performance.