云计算中融入贪心策略的调度算法研究  

Incorporate Greedy Strategy into Scheduling Algorithm for Cloud Computing

作　　者：周舟[1] 胡志刚[1] 

机构地区：[1]中南大学软件学院,长沙410004

出　　处：《小型微型计算机系统》2015年第5期1024-1027,共4页Journal of Chinese Computer Systems

基　　金：国家自然科学基金项目(61272148;60970038)资助

摘　　要：科学计算、商业和Web应用导致人们对计算力的需求越来越高,而现有数据中心的资源利用率普遍偏低.因此,在云计算环境中,合理的分配任务、实现最佳的调度极其必要.针对云计算中Min-Min算法优先调度小任务,而Max-Min算法优先调度大任务而导致负载不均衡的问题,提出一种算法即Min-Max.该算法对时间贪心,将小任务和大任务"捆绑"在一起执行调度,从而有效地解决了负载不均衡的问题.实验表明:Min-Max与Min-Min算法相比,提高了系统整体资源利用率,在任务总执行时间上节约了9%;Min-Max与Max-Min相比,除提高了系统整体利用率之外,在任务总体完成时间、平均任务响应时间上分别节约了7%和9%.Scientific,business and web-applications lead to the demand for computational power at a high level. However, utilization rate of resources in modern data centers is of a low level. Therefore, it is extremely urgent to distribute tasks to achieve the optimal scheduling scheme and complete computing tasks effectively. In view of Min-Min algorithm prefers scheduling small tasks and Max-Min algorithm prefers scheduling big tasks led to problem of load imbalance in cloud computing, a new algorithm named Min-Max is proposed. Min-Max makes good use of time for greedy strategy, small tasks and big tasks are put together for scheduling in order to solve the problem of load imbalance. Experimental results show that the Min-Max improves the utilization rate of entire system and saves 9 % overall execution time compared with Min-Min;as compared with Max-Min, Min-Max improves the utilization rate of entire system,the total completion time and average response time are saved 7% and 9% respectively.

关 键 词：云计算 贪心策略 负载均衡 MIN-MAX MIN-MIN MAX-MIN 

分 类 号：TP338[自动化与计算机技术—计算机系统结构]