Book Image

Deployment with Docker

By : Srdjan Grubor
Book Image

Deployment with Docker

By: Srdjan Grubor

Overview of this book

Deploying Docker into production is considered to be one of the major pain points in developing large-scale infrastructures, and the documentation available online leaves a lot to be desired. With this book, you will learn everything you wanted to know to effectively scale your deployments globally and build a resilient, scalable, and containerized cloud platform for your own use. The book starts by introducing you to the containerization ecosystem with some concrete and easy-to-digest examples; after that, you will delve into examples of launching multiple instances of the same container. From there, you will cover orchestration, multi-node setups, volumes, and almost every relevant component of this new approach to deploying services. Using intertwined approaches, the book will cover battle-tested tooling, or issues likely to be encountered in real-world scenarios, in detail. You will also learn about the other supporting components required for a true PaaS deployment and discover common options to tie the whole infrastructure together. At the end of the book, you learn to build a small, but functional, PaaS (to appreciate the power of the containerized service approach) and continue to explore real-world approaches to implementing even larger global-scale services.
Table of Contents (18 chapters)
Title Page
Credits
About the Author
Acknowledgments
About the Reviewer
www.PacktPub.com
Customer Feedback
Preface
Index

Docker image internals


To understand even better why we need persistent data, we first need to understand in detail how Docker handles container layers. We covered this topic in some detail in previous chapters, but here, we will spend some time to understand what is going on under the covers. We will first discuss what Docker currently does for handling the written data within containers.

How images are layered

As we covered earlier, Docker stores data that composes the images in a set of discrete, read-only filesystem layers that are stacked on top of each other when you build your image. Any changes done to the filesystem are stacked like transparent slides on top of each other to create the full tree, and any files that have newer content (including being completely removed) will mask the old ones with each new layer. Our former depth of understanding here would probably be sufficient for the basic handling of containers, but for advanced usage, we need to know the full internals on how...