Book Image

Mastering Kubernetes - Fourth Edition

By : Gigi Sayfan
3.3 (3)
Book Image

Mastering Kubernetes - Fourth Edition

3.3 (3)
By: Gigi Sayfan

Overview of this book

The fourth edition of the bestseller Mastering Kubernetes includes the most recent tools and code to enable you to learn the latest features of Kubernetes 1.25. This book contains a thorough exploration of complex concepts and best practices to help you master the skills of designing and deploying large-scale distributed systems on Kubernetes clusters. You’ll learn how to run complex stateless and stateful microservices on Kubernetes, including advanced features such as horizontal pod autoscaling, rolling updates, resource quotas, and persistent storage backends. In addition, you’ll understand how to utilize serverless computing and service meshes. Further, two new chapters have been added. “Governing Kubernetes” covers the problem of policy management, how admission control addresses it, and how policy engines provide a powerful governance solution. “Running Kubernetes in Production” shows you what it takes to run Kubernetes at scale across multiple cloud providers, multiple geographical regions, and multiple clusters, and it also explains how to handle topics such as upgrades, capacity planning, dealing with cloud provider limits/quotas, and cost management. By the end of this Kubernetes book, you’ll have a strong understanding of, and hands-on experience with, a wide range of Kubernetes capabilities.
Table of Contents (21 chapters)
19
Other Books You May Enjoy
20
Index

High-availability, scalability and capacity planning

Highly available systems must also be scalable. The load on most complicated distributed systems can vary dramatically based on time of day, weekday vs weekend, seasonal effects, marketing campaigns and many other factors. Successful systems will have more users over time and accumulate more and more data. That means that physical resources of the clusters - mostly nodes and storage - will have to grow over time too. If your cluster is under provisioned it will not be able to satisfy all the demand and it will not be available because requests will time out or be queued up and not processed fast enough.

This is the realm of capacity planning. One simple approach is to over-provision your cluster. Anticipate the demand and make sure you have enough of a buffer for spikes of activity. But, this approach suffers from several deficiencies:

  • For highly dynamic and complicated distributed systems it’s difficult to predict the demand...