Computer Science and Engineering, Bangladesh Army International University of Science and Technology, Cumilla, Bangladesh.
Computer Science and Engineering, International University of Business Agriculture and Technology, Dhaka, Bangladesh.
PLoS One. 2024 Aug 15;19(8):e0304517. doi: 10.1371/journal.pone.0304517. eCollection 2024.
Cloud computing is a popular, flexible, scalable, and cost-effective technology in the modern world that provides on-demand services dynamically. The dynamic execution of user requests and resource-sharing facilities require proper task scheduling among the available virtual machines, which is a significant issue and plays a crucial role in developing an optimal cloud computing environment. Round Robin is a prevalent scheduling algorithm for fair distribution of resources with a balanced contribution in minimized response time and turnaround time. This paper introduced a new enhanced round-robin approach for task scheduling in cloud computing systems. The proposed algorithm generates and keeps updating a dynamic quantum time for process execution, considering the available number of process in the system and their burst length. Since our method dynamically runs processes, it is appropriate for a real-time environment like cloud computing. The notable part of this approach is the capability of scheduling tasks with asymmetric distribution of burst time, avoiding the convoy effect. The experimental result indicates that the proposed algorithm has outperformed the existing improved round-robin task scheduling approaches in terms of minimized average waiting time, average turnaround time, and number of context switches. Comparing the method against five other enhanced round robin approaches, it reduced average waiting times by 15.77% and context switching by 20.68% on average. After executing the experiment and comparative study, it can be concluded that the proposed enhanced round-robin scheduling algorithm is optimal, acceptable, and relatively better suited for cloud computing environments.
云计算是现代世界中一种流行、灵活、可扩展且具有成本效益的技术，它提供按需动态服务。用户请求的动态执行和资源共享设施要求在可用的虚拟机之间进行适当的任务调度，这是一个重要问题，在开发优化的云计算环境中起着关键作用。轮询是一种常见的调度算法，用于公平分配资源，并在最小化响应时间和周转时间方面做出平衡贡献。本文提出了一种新的增强型轮询方法，用于云计算系统中的任务调度。所提出的算法生成并不断更新用于进程执行的动态量子时间，考虑到系统中可用的进程数量及其突发长度。由于我们的方法可以动态运行进程，因此它适合于云计算等实时环境。该方法的显著特点是能够调度具有非对称突发时间分布的任务，避免了车队效应。实验结果表明，与现有的改进轮询任务调度方法相比，该算法在最小化平均等待时间、平均周转时间和上下文切换次数方面表现更好。与其他五种增强型轮询方法进行比较，该方法的平均等待时间减少了 15.77%，上下文切换减少了 20.68%。通过执行实验和比较研究，可以得出结论，所提出的增强型轮询调度算法是最优的、可接受的，并且相对更适合云计算环境。
