DRAPS: Dynamic and resource-aware placement scheme for docker containers in a heterogeneous cluster

Ying Mao, Jenna Oak, Anthony Pompili, Daniel Beer, Tao Han, Peizhao Hu

2017 · 96 citations · 22 references

Concepts

TL;DR

Virtualization has enabled cloud computing, but virtual machines suffer from redundant boot volumes and slow provisioning, leading to the adoption of containers such as Docker, yet Docker Swarm does not account for heterogeneity in node resources and service demands. This study aims to develop DRAPS, a resource‑aware placement scheme that improves performance in heterogeneous Docker container clusters. DRAPS dynamically assigns containers to nodes by matching their CPU and memory requirements with node capacities, thereby optimizing resource utilization and reducing bottlenecks.

Abstract

Virtualization is a promising technology that has facilitated cloud computing to become the next wave of the Internet revolution. Adopted by data centers, millions of applications that are powered by various virtual machines improve the quality of services. Although virtual machines are well-isolated among each other, they suffer from redundant boot volumes and slow provisioning time. To address limitations, containers were born to deploy and run distributed applications without launching entire virtual machines. As a dominant player, Docker is an open-source implementation of container technology. When managing a cluster of Docker containers, the management tool, Swarmkit, does not take the heterogeneities in both physical nodes and virtualized containers into consideration. The heterogeneity lies in the fact that different nodes in the cluster may have various configurations, concerning resource types and availabilities, etc., and the demands generated by services are varied, such as CPU-intensive (e.g. Clustering services) as well as memory-intensive (e.g. Web services). In this paper, we target on investigating the Docker container cluster and developed, DRAPS, a resource-aware placement scheme to boost the system performance in a heterogeneous cluster.

References

22