A Container Deployment Optimization Method for Edge Computing

Abstract

In recent years, with the rapid development of Cloud Computing, Big Data and Internet of Things, the amount of data generated by edge nodes grows profusely, which results in high network bandwidth requirements. In order to improve the QoS of Cloud Computing, edge computing has received wide-spread attention in both industry and academia. As a resource isolation technology for edge nodes, container virtualization has been widely used. However, under the condition of unstable network connections in edge environments, there are some shortcomings on native docker container, such as large amount of image data transmission, network interruption and retransmission cost, which affect the quality of edge services. Therefore, this paper proposes an Images Distribution Method based on Deduplication (IDM-Dedup), which improves the transmission efficiency of container image and reduces the time to deploy and migrate edge services. The experimental results show that IDM-Dedup is more effective than native docker container method.