Container oriented job scheduling using linear programming model

Abstract

With Docker gaining widespread popularity in the recent years, the container scheduler becomes a crucial role for the exploding containerized applications and services. In this work, the container host energy conservation, the container image pulling costs from the image registry to the container hosts and the workload network transition costs from the clients to the container hosts are evaluated in combination. By modeling the scheduling problem as an integer linear programming, an effective and adaptive scheduler is proposed. Impressive cost savings were achieved compared to Docker Swarm scheduler. Moreover, it can be easily integrated into the open-source container orchestration frameworks.