论文部分内容阅读
针对混合云的功能和复杂度逐渐增大而导致资源失效率增高的问题,提出一种提高混合云资源调度成功率的调度算法.首先,根据私有云资源失效规律特点,建立资源有效利用率模型和任务稳定性模型.然后综合考虑公共云和私有云的特点,建立基于失效感知的两层资源调度模型(2L-FARS),并使用建立的ST—LLF(任务稳定性阀值控制的最低松弛度优先调度算法)和DQPA(双队列资源提供算法)调度算法分别完成两层资源调度.最后使用failure traces和workload traces,对提出的策略进行验证.实验结果表明,该策略有效地减少了任务截止期违约率,并且在提高资源利用率的同时,一定程度上降低了任务执行总费用.
Aiming at the problem that the function and complexity of hybrid cloud increase, the efficiency of resource failure increases, a scheduling algorithm is proposed to improve the success rate of hybrid cloud resource scheduling.Firstly, according to the characteristics of failure rules of private cloud resources, an efficient resource utilization model And task stability model.Then considering the characteristics of public cloud and private cloud, a two-layer Resource Scheduling Model (2L-FARS) based on failure perception is established and established using the established ST-LLF (minimum relaxation of task stability threshold control Degree scheduling algorithm) and DQPA (dual queue resource providing algorithm) scheduling algorithm to complete the two-tier resource scheduling respectively.Finally, the use of failure traces and workload traces to validate the proposed strategy.The experimental results show that the strategy effectively reduces the task deadline Term default rate, and improve the utilization of resources at the same time, to a certain extent, reduced the total cost of task execution.