Docker for Java Developers

Preface

The Java programming language was created over 20 years ago. It continues to be the most popular and widely used programming language after all these years. The design patterns and antipatterns of Java deployment are well known. The usual steps to deploy a Java application involve using a script that downloads and installs the operating system package such as JDK on a machine—whether physical or virtual. Operating system threads and memory need to be configured, the network needs to be set up, the correct database identified, and several other such requirements need to be configured for the application to work. These applications are typically deployed on a virtual machine (VM). Starting up these VMs is an expensive operation and can take quite a few minutes in most cases. The number of VMs that can run on a host is also limited because the entire operating system needs to be started, and thus there are stringent requirements on CPU and memory of the host.

Containers provide several benefits over traditional VM-based deployments. Faster startup and deployments, security and network sandboxing, higher density, and portability across different environments are some of the commonly known advantages. They also improve portability across machines and reduce the impedance mismatch between dev, test, and prod environments.

There are efforts like the Open Container Initiative (OCI) that aim to create an industry standard around container formats and runtime. Docker is the first ...