Horizontal Pod Autoscaling in Kubernetes for Elastic Container Orchestration.

Thanh-Tung Nguyen,Dae-Heon Park,Yu-Jin Yeom,Sehan Kim,Taehong Kim

doi:10.3390/s20164621

Abstract

Kubernetes, an open-source container orchestration platform, enables high availability and scalability through diverse autoscaling mechanisms such as Horizontal Pod Autoscaler (HPA), Vertical Pod Autoscaler and Cluster Autoscaler. Amongst them, HPA helps provide seamless service by dynamically scaling up and down the number of resource units, called pods, without having to restart the whole system. Kubernetes monitors default Resource Metrics including CPU and memory usage of host machines and their pods. On the other hand, Custom Metrics, provided by external software such as Prometheus, are customizable to monitor a wide collection of metrics. In this paper, we investigate HPA through diverse experiments to provide critical knowledge on its operational behaviors. We also discuss the essential difference between Kubernetes Resource Metrics (KRM) and Prometheus Custom Metrics (PCM) and how they affect HPA’s performance. Lastly, we provide deeper insights and lessons on how to optimize the performance of HPA for researchers, developers, and system administrators working with Kubernetes in the future.

Highlights

In recent years, with the rapid emergence of cloud computing and later edge computing, virtualization techniques have become a sensational topic for both academic research and industrial development as they enable Cloud platforms such as Amazon Web Services (AWS) [1], Google CloudPlatform (GCP) [2], Microsoft Azure [3] to achieve elasticity on a large scale [4]
One of the emerging virtualization techniques is containerization technology, in which a lightweight operating system (OS) equipped with ready-to-deploy application components is packaged into a self-sufficient container ready to run on a host machine that supports multi-tenancy [4,5]
As containers can be deployed on a large scale [7], there is a tremendous need for container orchestration platforms that are highly automatic in terms of deployment, scaling, and management

Summary

Introduction

Platform (GCP) [2], Microsoft Azure [3] to achieve elasticity on a large scale [4]. One of the emerging virtualization techniques is containerization technology, in which a lightweight operating system (OS) equipped with ready-to-deploy application components is packaged into a self-sufficient container ready to run on a host machine that supports multi-tenancy [4,5]. Amongst various orchestration platforms including Docker Swarm [8], Amazon Elastic Container. Service (Amazon ECS) [9], Red Hat OpenShift Container Platform (Red Hat OCP) [10], Kubernetes [11]. Has become the de facto standard for its popularity It is an open-source platform, on which it is easy to package and run containerized applications, workloads, and services, and provides a framework for operating scalable distributed systems.

Objectives

Discussion

Conclusion