【摘要】 针对Docker动态迁移中高脏页率内存被重复拷贝问题,提出了基于Markov的Docker动态迁移优化算法。首先利用linux的内存跟踪机制获取内存页被修改的情况,之后将获取的数据输入Markov预测算法,根据预测结果求出高脏页率内存并将其放到后期传输。实验表明,该方案能有效减少高脏页率内存的重复拷贝,特别是在高负载的情况下,同时能减少迭代次数约25.20%,缩短停机时间约7.14%。更多还原

【Abstract】 To solve the problem that high dirty page rate memory is copied repeatedly in Docker dynamic migration,an optimization algorithm of Docker dynamic migration based on Markov is proposed.Firstly,the memory tracking mechanism of Linux is used to obtain the situation that the memory page is modified.Then,the acquired data is input into Markov prediction algorithm,and according to the prediction results,the memory with high dirty page rate is calculated and put into the later transmission.Experimental results show that this scheme can effectively reduce the repeated copies of memory with high dirty page rate,especially under the condition of high load,at the same time,it can reduce the number of iterations by about 25.20% and shorten the downtime by about 7.14%.更多还原