Dynamic resource allocation for an energy efficient VM architecture for cloud computing

Minimizing power consumption is a vital consideration in the modern-day development of cloud computing. One of the major challenges reported in cloud computing is the consumption of power by computing resources due to improper allocation of resources over improperly sized virtual machines (VM). In spite of many efforts, the existing solutions are only able to meet the requirement for minimizing power consumption to a limited extent, due to their lack of optimized allocation of computing resources. The primary aim of the proposed work is to make effective use of the computing resources of the cloud for minimizing power consumption. It employs the concept of mapping appropriately sized VMs to a group of tasks in a data center, in order to reduce its power consumption. It involves the clustering of tasks on the basis of their computing requirements and finding a suitably sized VM with the required computing resources. The efficient use of computing resources on the basis of their actual requirements for a group of tasks helps to save a substantial amount of power. The proposed work is evaluated for its superiority over representational techniques using Google cloud traces as benchmark dataset. The experimental results showed an improvement of 8.42% in power consumption compared to representational techniques using fixed-sized VMs in the field. The proposed approach also achieves an improvement of 62% in the number of instances of VMs created for hosting the task workload, while maintaining a low task rejection rate