Threshold Based Load Balancing Algorithm in Cloud Computing
Threshold Based Load Balancing Algorithm in Cloud Computing
复制标题
DOI：
10.1109/jcc56315.2022.00011
复制
发表时间：
2022-08
期刊：
影响因子：
--
通讯作者：
Shusmoy Chowdhury;Ajay Katangur
中科院分区：
文献类型：
--
作者：
Shusmoy Chowdhury;Ajay Katangur
云计算已成为软件行业的新兴趋势，需要大量的基础设施和资源。云计算未来的成功取决于基础设施实例化的有效性和可用资源的利用。负载均衡可确保满足这些条件，从而改善用户的云环境。负载平衡在节点之间动态分配工作负载，确保单个资源不会因任务而不堪重负或未得到充分利用。在本文中，我们提出了一种基于阈值的负载平衡算法，以确保节点之间工作负载的平均分配。该算法的主要目标是当有活动任务时，阻止云中的虚拟机因任务过载或因任务分配不足而空闲。我们在 Cloudanalyst 模拟器中使用真实世界的数据场景模拟了我们提出的算法。仿真结果表明，与先来先服务（FCFS）、循环（RR）和等分当前执行负载平衡算法（ESCELB）等现有算法相比，我们提出的基于阈值的算法可以为数据中心提供更好的任务/请求响应时间和数据处理时间。
Cloud computing has become an emerging trend for the software industry with the requirement of large infrastructure and resources. The future success of cloud computing depends on the effectiveness of instantiation of the infrastructure and utilization of available resources. Load Balancing ensures the fulfillment of these conditions to improve the cloud environment for the users. Load Balancing dynamically distributes the workload among the nodes in such a way that no single resource is either overwhelmed with tasks or underutilized. In this paper we propose a threshold based load balancing algorithm to ensure the equal distribution of the workload among the nodes. The main objective of the algorithms is to stop the VMs in the cloud being overloaded with tasks or being idle for lack allocation of tasks, when there are active tasks. We have simulated our proposed algorithm in the Cloudanalyst simulator with real world data scenarios. Simulation results shows that our proposed threshold based algorithm can provide a better response time for the task/requests and data processing time for the datacenters compared to the existing algorithms such as First Come First Serve (FCFS), Round Robin(RR) and Equally Spread Current Execution Load Balancing algorithm(ESCELB).