Recent advances in Container Orchestration: Network-Aware Scheduling in Container Clouds

Abstract

Containers have revolutionized application deployment and life-cycle management in current cloud platforms. Applications have evolved from single monoliths to complex graphs of loosely-coupled microservices aiming to improve deployment flexibility and operational efficiency. However, the efficient orchestration of containerized applications is challenging due to their complex inter-dependencies. Further, recent applications are becoming even more delay-sensitive, demanding lower latency between dependent microservices. Scheduling policies in popular container orchestration platforms mainly aim to increase the resource efficiency of the infrastructure, insufficient for these latency-sensitive applications. Application domains such as the Internet of Things (IoT) and multi-tier web services would benefit from network-aware policies that consider network latency and bandwidth in the scheduling process. This tutorial provides an overview of application scenarios and methodologies to address the efficient orchestration of containerized applications. After that, the tutorial provides a practical vision of network-aware scheduling via the developed Diktyo framework 1,2 for the popular Kubernetes (K8s) platform. Diktyo determines the placement of dependent microservices in long-running applications focused on reducing the application’s end-to-end latency and guaranteeing bandwidth reservations. The tutorial includes a live-demo showing the benefits of the Diktyo framework by deploying typical containerized applications. The framework has been open-sourced and already accepted in the K8s scheduling community repository as an alternative scheduler. We want to showcase to attendees how they could apply Diktyo to deploy their applications with network awareness in a K8s cluster. Lastly, a selected set of lessons learned within this dedicated area of container management is presented, and future trends in container scheduling are outlined.