A particle swarm optimization-based container scheduling algorithm of docker platform

Abstract

Docker is rapidly changing the rules of operation in the field of cloud computing, and completely subverts the development of cloud technology. Swarm is a Docker container-based cluster management tool. By analyzing and researching the overall architecture and scheduling strategy of Swarm, in this paper, we propose a Particle Swarm Optimization-based container scheduling algorithm of Docker platform, which to have solved the problem of insufficient resource utilization and load balance. We distribute application containers on Docker hosts, balance resource usage, and ultimately improve application performance. Experimental results show that the performance of the PSO is improved 20% and 19% higher than the spread and improved random respectively under the same host configuration.