Multi-Algorithm Collaboration Scheduling Strategy for Docker Container
Qilong Li, Yu Long Fang · 2017 International Conference on Computer Systems, Electronics and Control (ICCSEC) · 2017
As the most popular container technology, Docker can provide more efficient solution for virtualization. Swarm is a management tool for Docker cluster. Since it is simple and flexible, Swarm is suitable for building private cloud platform. However, the strategies Swarm currently supports cannot achieve good performance of load balancing in some cases. To solve this problem, we present a new scheduling strategy named Multi-Algorithm Collaboration Scheduling Strategy, which can satisfy various scheduling requirements by using different algorithm specifically designed for different cases to process requests in turns. The results of comparison experiment suggest that the presented strategy has a more comprehensive performance than other strategies for Swarm.