A Systematic Resource Management for VR Streaming on MECs

Mobile edge computing(MEC) has brought computation and storage resources from the centralized data center to the edge network; and, this could effectively reduce end-to-end latency for VR users. In this work, we implement a VR streaming and its resource management system based on Kubernetes and KubeVirt for VR streaming on MECs. Such system is able to adjust the resource pool according to user requirements and recommend a users with proper VM template based on historical logs. Moreover, our system could let users run VR applications on remote servers, and stream their contents to the user's VR headsets. At last, the system administrator could control through the web interface and view the monitoring of resource usage in real time. ? 2021 IEEE.