论文部分内容阅读
负载调度是云计算得以大规模应用及提高服务性能的关键技术,对提高云供应商服务质量、用户满意度以及数据中心集群资源利用率等有极其重要的意义.云计算环境中,由于用户任务类型的不同,对带宽的需求也不尽相同,若不区分不同任务对不同带宽的要求,可能会造成资源的浪费,增长用户等待时间.本文对经典Min-Min算法进行改进,提出了BCLL-Min-Min算法,该算法满足带宽需求约束,并且实现相对负载均衡调度.仿真实验表明,BCLL-Min-Min算法能够适应云计算环境下任务多样性和不确定性的情况,使用该调度算法可以提高集群的吞吐率、较大改善数据中心的负载均衡性.
Load scheduling is a key technology for cloud computing to be applied on a large scale and improve service performance, which is of great significance to improve service quality, user satisfaction and data center cluster resource utilization, etc. In cloud computing environment, user tasks Different types require different bandwidths, and if different tasks are not required, the resources may be wasted and the waiting time may be increased. In this paper, the classical Min-Min algorithm is improved and BCLL- Min-Min algorithm, which meets the bandwidth requirements constraints and achieves relative load balancing scheduling.The simulation results show that the BCLL-Min-Min algorithm can adapt to the diversity and uncertainty of tasks under the cloud computing environment, and the scheduling algorithm can be used Improve the throughput of the cluster, greatly improve the data center load balancing.