云计算环境下虚拟机在线迁移优化

在线迁移在云计算平台下已广泛使用,虚拟机内存迁移主要采用的是预拷贝算法.传统预拷贝算法在迭代过程中会将脏页反复重传导致迁移时间较长,针对这一问题文章提出了在拷贝过程中增加脏页预测算法和内存压缩算法相结合的方法.脏页预测算法是采用统计方法,依据内存页历史操作情况对改动频繁的内存页进行标记,根据标记决定是否在最后一轮内存迁移中传输,为了减少迁移传输量,先通过内存压缩算法将其压缩,然后再传输.实验结果表明,改进后的方法有效地降低了停机时间和迁移时间,提高了迁移效率,达到了更快迁移的目的.

Live migration optimization for virtual machine in cloud computing environment

Live migration of VM has been widely used in cloud computing platforms, migration of the memory mainly uses the pre-copy algorithm. Memory pages repeated retransmission causing longer time in the traditional process of memory pre-eopy, this paper proposed a method of adding the dirty page prediction algorithm and memory compression algorithm in the process of the pre-copy. By judging these frequently updated pages using the dirty pages history of the operation visits, decide whether put the page into the new bitmap and transmitted in the last round of the iteration proeessonly. To reduce the migration of traffic, through the compression algorithm, the pages are compressed and then are transmited. The experimental results show that, the improved algorithm significantly reduces the migration time and downtime, improved the efficiency of the live migration, reached the goal of faster migrating the virtual machine.