Live migration destination selecting method for disaster recovery on e-learning environment
Satoshi Togawa, Kazuhide Kanenishi · 2016
In this research, we have made a study how to decide the live-migration destination on our proposed disaster recovery framework for e-Learning environment. Our main research is to build a framework of disaster recovery such as against earthquake, tsunami and heavy floods. Especially, our proposed framework is based on private cloud collaboration. We build a prototype system based on IaaS architecture, and this prototype system is constructed by several private cloud fabrics on each site such as several universities. The virtual machine that drives a LMS (Learning Management System) such as Moodle is operated on the private cloud fabrics. When the private cloud fabric is affected by the natural disaster, these virtual machines which are operated on the private cloud fabric are migrated to other private cloud fabric safer than affected site. Therefore, we have to make a method to select safer destination for live-migration operation. We use the information such as earthquake alert, and the weather information like amount of rainfall. The live-migration destination is decided by these information for keeping e-Learning environment. In this paper, we show our private cloud framework. And, we describe our decision method to select live-migration destination. Then, we show the experimental results on the prototype system.