当前位置: X-MOL 学术Wireless Pers. Commun. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
QoS Enhancement in Cloud-IoT Framework for Educational Institution with Task Allocation and Scheduling with Task-VM Matching Approach
Wireless Personal Communications ( IF 1.9 ) Pub Date : 2021-07-01 , DOI: 10.1007/s11277-021-08634-6
Sunil Kumar Chowdhary , A. L. N. Rao

The Cloud-IoT framework offers on-demand service for numerous applications with the aid of data gathered by IoT and the computing resources of cloud computing. The quality of service (QoS) degrades due to task-VM mismatch due to the heterogeneous service request from IoT devices. The tasks processed by an inappropriate VM may cause delay and affect the Quality of Service (QoS). The proposed task allocation and scheduling algorithm aim is to improve the QoS of education service offered by Cloud-IoT in an educational organisation. In the task allocation stage, task VM pairs are prioritized initially and task-VM pairs are selected based on the minimum of the expected completion time (ECT) with the approach named Priority Based Task Allocation and Buffering (PBTAB) Algorithm. In this stage, at each of the clouds, the selected task-VM pairs are placed on queues based on the proximal value of the MCT. In the scheduling stage, task-VM pair matching (T-VMBS) Algorithm schedules the task with the selection of the best of the VM from the total clouds to speed up the task execution. The PBTAB and T-VMBS algorithm achieved throughput performance of more than 90% with larger dataset and huge number of VM. The proposed approach achieved a decreased makespan of less than 50%. Similarly deadline violation rate and average reliability exhibited a better performance.



中文翻译:

使用 Task-VM 匹配方法进行任务分配和调度的教育机构 Cloud-IoT 框架中的 QoS 增强

Cloud-IoT 框架借助 IoT 收集的数据和云计算的计算资源为众多应用程序提供按需服务。由于来自 IoT 设备的异构服务请求,任务-VM 不匹配导致服务质量 (QoS) 下降。不合适的虚拟机处理的任务可能会导致延迟并影响服务质量(QoS)。所提出的任务分配和调度算法旨在提高教育组织中 Cloud-IoT 提供的教育服务的 QoS。在任务分配阶段,首先对任务 VM 对进行优先级排序,并使用称为基于优先级的任务分配和缓冲 (PBTAB) 算法的方法根据预期完成时间 (ECT) 中的最小值选择任务 VM 对。在这个阶段,在每一朵云,选定的任务-VM 对根据 MCT 的近端值放置在队列中。在调度阶段,任务-VM对匹配(T-VMBS)算法从总云中选择最佳VM来调度任务以加快任务执行。PBTAB 和 T-VMBS 算法在更大的数据集和大量的 VM 下实现了 90% 以上的吞吐量性能。所提出的方法实现了小于 50% 的缩短完工时间。同样,最后期限违反率和平均可靠性表现出更好的性能。PBTAB 和 T-VMBS 算法在更大的数据集和大量的 VM 下实现了 90% 以上的吞吐量性能。所提出的方法实现了小于 50% 的缩短完工时间。同样,最后期限违反率和平均可靠性表现出更好的性能。PBTAB 和 T-VMBS 算法在更大的数据集和大量的 VM 下实现了 90% 以上的吞吐量性能。所提出的方法实现了小于 50% 的缩短完工时间。同样,最后期限违反率和平均可靠性表现出更好的性能。

更新日期:2021-07-01
down
wechat
bug