Docker, and containers in general, are very powerful tools worth digging into. By combining resource isolation features, including union capable filesystems, Docker allows for the creation of packages called containers that include everything needed to run an application. Containers, like virtual machines, are self-contained, but instead of virtualizing the hardware, they virtualize the operating system itself. In practice, this makes a huge difference. As you have probably noticed by now, starting a virtual machine, such as an EC2 instance, takes time. This comes from the fact that, in order to start a virtual machine, the hypervisor (that's the name of the technology that creates and runs virtual machines) has to simulate all the motions involved in starting a physical server, loading an operating system, and finally, going through the different run-levels. In addition, virtual machines have a much larger footprint on disk and in memory. With Docker, the added layer is hardly noticeable, and the size of the containers can stay very small. In order to better illustrate that, we will first install Docker and explore its basic usage a bit.