基于滚动优化的密码云实时任务阈值调度方法

发布时间：2019-07-08 18:15

【摘要】：针对当前云任务调度算法在密码云环境中无法实现任务实时处理的问题,提出一种基于滚动优化窗口的实时阈值调度方法。首先,将密钥调用环节融入密码任务流程中,提出一种密码云服务架构;其次,为实现任务的实时调度,构建基于滚动窗口的密码任务调度器模型和吞吐量分析模型,用于获得实时的吞吐量数据;最后,为满足云租户对高速密码服务的客观需求,提出吞吐量阈值调度算法,从而根据实时吞吐量相对于吞吐量阈值的变化情况实时迁移虚拟密码机。仿真结果表明,该方法与未采用滚动优化窗口或虚拟机迁移技术的方法相比,具有任务完成时间短、CPU占用率低的特点,且实时吞吐量能够持续保持在网络带宽的70%~85%,从而验证了其在密码云环境中的有效性和实时性。
[Abstract]:In order to solve the problem that the current cloud task scheduling algorithm can not realize the real-time task processing in the password cloud environment, a real-time threshold scheduling method based on rolling optimization window is proposed. Firstly, the key call link is integrated into the password task flow, and a password cloud service architecture is proposed. Secondly, in order to realize the real-time task scheduling, the password task scheduler model and throughput analysis model based on rolling window are constructed to obtain real-time throughput data. Finally, in order to meet the objective needs of cloud tenants for high-speed cryptography services, a throughput threshold scheduling algorithm is proposed to migrate virtual cryptography in real time according to the change of real-time throughput relative to throughput threshold. The simulation results show that compared with the method without rolling optimization window or virtual machine migration technology, this method has the characteristics of short task completion time, low CPU utilization rate, and the real-time throughput can be kept at 70% 鈮，

本文编号：2511780