A hybrid approach for load balancing in cloud computing

Abstract

Cloud computing is one of the most promising technical developments in recent days. It emerged as a dominant and transformational paradigm. Cloud computing can be considered a vital form of information technology that allows the delivery of services to users via the Internet upon request from the user and based on immediate payment. One of the main challenges and important fields for research in the cloud computing environment is load balancing. Load balancing has become an important point for stability and good system performance. Therefore, the main goal is to establish an effective load balancing algorithm for task scheduling. In this paper, we introduce the Hybrid Algorithm for Load Balancing (HALB), which aims to balance effective loading among virtual machines by balancing the percentage of data usage from RAM in each VM. As the results of the percentages were close, and all percentages did not reach the condition of overload. The proposed hybrid algorithm also reduces average waiting time and turnaround time. As a mechanism of work of our hybrid algorithm relies on two types of scheduling algorithms, one dependent on the other, namely the lottery algorithm and the Shortest Job First algorithm. A specific mechanism has been implemented to allocate tasks resulting from scheduling each algorithm separately, when calculating the total size of data tasks in each VM, the results showed that the volume of data allocated within the VM. Data sizes converge across all VMs.