An Efficient Packet Priority Scheduling Algorithm for IO Virtualization of Cloud

摘要

Virtualization has a considerable role in the cloud computing where the multiple users intend to share multiple resources. It improves resource availability and provides cost-effectiveness and high flexibility. The driver domain model cannot perform well in the virtualization process where the data sharing becomes the complex issue. Virtual Machine becomes the vital component in the networking environment where it is responsible for receiving and transmitting the data packets with the consideration of priority and fairness. The worst performance of virtual machine leads the degraded networking performance. This problem is overcome in the proposed methodology by proposing a Fuzzy Hierarchical Cluster Based Round Robin Priority Scheduling Algorithm. In the proposed scheme, each VM, except those at the last level of the virtual hierarchy in the zone based VMs, has three levels of priority cluster. First, we present a fuzzy hierarchical clustering algorithm for clustering packets. Then, we present an approach to construct fuzzy logic rules based on the packet clusters. Finally, we apply scheduling algorithm in which the clustered packets are scheduled between the VMs and driver domain to VM; it improves the efficient resource utilization, implementation complexity and scalability. The proposed algorithm works based on the priority based scheduling where the scheduling is done in the round robin fashion. The performance evaluation conducted were proves that the proposed methodology provides better result than the existing approaches in terms of improved turnaround time, waiting time and the number of context switches. The time complexity of the PRR algorithm is 0(1). From the experimental results shows that improve the efficient resource utilization, implementation complexity and scalability.