VM Migration Considering Downtime for Accuracy Improvement in Multi-stage Information Processing System
Kazutoshi Nakane, Takumi Anjiki, Jiquan Xie, Yukinobu Fukushima, Tutomu Murase · 2022 IEEE International Conference on Consumer Electronics - Taiwan · 2022
In this paper, we propose a VM migration method that improves the accuracy in edge nodes of multi-stage information processing system even when the VM migration downtime is large. The previous method only performs well with short VM migration times. In contrast to previous studies, we place VMs to make a large contribution to improving accuracy by considering the system load degree and the characteristics of the tasks generated by each VM. The evaluation results show that the proposed method improves the accuracy by about 68% compared to the conventional method when the number of VMs is 25.