Research and Application of Auto-Scaling Unified Communication Server Based on Docker
Yu Jin-Gang, Zhai Ya-Rong, Bo Yu, Li Shu · 2017
With the continuous development of communication technology, Unified Communication (UC) has also attracted much attention. At present, most of the Unified Communication servers are deployed on virtual machines, but virtual machines have the characteristics of large resource occupation, slow startup speed, poor flexibility and low efficiency, which brings some problems. In contrast, container technology is characterized by run-time lightweight and application deployment fast, combined with the characteristics of the container, this paper presents a design of auto-scaling Unified Communication servers based on Docker. This paper uses Kubernetes as a choreographer. As the current Kubernetes auto-scaling algorithm is poor flexible, we propose an auto-scaling mechanism. It combines the responsive scaling with the predictive scaling, the reliability of the system is ensured, greatly improves the flexibility of Unified Communication deployment, speeds up the introduction of new business speed, and has very strong elastic expansion capacity.