An Effective Approach on Data Visualization Using Cluster Graphs

Abstract

Load balancing, which distributes changing workloads and computing resources in a cloud context, is the main difficulty in cloud computing. The load balancing algorithm seeks to balance the total system load by moving workload from overloaded to underloaded nodes in order to ensure the system's overall performance. This work offers a Hybrid Dynamic Degree Balance (HDDB) algorithm for allocating virtual machines (VMs) to the best-suited host based on CPU availability and host membership value. The proposed scheduling hybridizes two algorithms namely Dynamic Degree Balance CPU Based (D2B_CPU based) and Dynamic Degree Balanced Membership based (D2B_Membership). The suggested algorithm HDDB was tested using the CloudSim simulation tool. To verify the performance of proposed algorithms, performance metrics used are turnaround time of cloudlets, execution cost, throughput time, degree of imbalance, CPU utilization, bandwidth utilization and memory utilization. The results show a considerable improvement in performance when compared to other current algorithms such as Dynamic Degree Balance CPU Based (D2B CPU), Dynamic Degree Balanced Membership Based (D2B Membership), Round Robin (RR), First Come First Serve (FCFS), and Shortest Job First (SJF).