Docker is a powerful tool for creating and deploying applications. It simplifies rolling out applications across multiple systems and is a useful tool for integrating new technologies. An application that runs using Docker will start up the same every time on every system. This means that if the application works on your local computer, it’ll work anywhere that supports Docker. That’s great news! It simplifies your development process and can be a powerful tool for continuous delivery.
As you begin to understand Docker, you need to grasp two key facets of how Docker works. Docker image vs container — it can be a little bit complicated at first. In this post, we break them down and make them easy to understand.
What are images?
If you’ve ever used virtual machines aside from Docker, you’ll be familiar with Docker images. In other virtual machine environments, images would be called something like “snapshots.” They’re a picture of a Docker virtual machine at a specific point in time. Docker images are a little bit different from a virtual machine snapshot, though. For starters, Docker images can’t ever change. Once you’ve made one, you can delete it, but you can’t modify it. If you need a new version of the snapshot, you create an entirely new image.
This inability to change (called “immutability”) is a powerful tool for Docker images. An image can never change. So, if you get your Docker virtual machine into a working state and create an image, you know that image will always work, forever. This makes it easy to try out additions to your environment. You might experiment with new software packages, or try to reorganize your project files. When you do this, you can be sure that you won’t break your working instance, because you can’t. You will always be able to shut down your Docker virtual machine and restart it using your existing image, and it’ll be like nothing ever changed.
What’s a container?
We use the phrase “Docker virtual machine,” but the better way to say that is “Docker container.” If a Docker image is a digital photograph, a Docker container is like a printout of that photograph. In technical terms, we call it an “instance” of the image. Each Docker container runs separately, and you can modify the container while it’s running. Modifications to a Docker container aren’t saved unless you create another image, as we noted. Most Docker images include full operating systems to allow you to do whatever you need on them. This makes it easy to start up a program—like a command line—on the running container. Inside that command line, you can do some work like installing a new software package or configuring the system’s security. Then you can save another image and upload it to somewhere like Docker Hub to share it with people who can make use of your work.
Sometimes, you need to do things on a container that you need to save, but can’t become part of the Docker image. A good example is building a web application, where you’ll probably have a Docker container that holds your database. Databases need to be able to write data to the hard drive so that it can be retrieved later. Docker has the ability to configure containers and “share” folders between the container and the host computer. One of the most common use cases for this is to share a directory that holds your application code. You modify the application code on your host machine, and those changes are detected by the application server running inside the Docker container.
Containers and images work together
Docker containers and images work together to unlock the potential of Docker. Each image provides an infinitely reproducible virtual environment shareable across the room or around the world. Containers build on those images to run applications—both simple or very complicated. What’s more, some terrific tools like Docker Compose make it simple to “compose” novel Docker systems encompassing multiple containers using a small config file. You can easily use images for a database, web server, caching server and message queue to easily configure a web application, for instance. Once all of those pieces come together, monitoring them with a Docker-aware application monitoring platform like Retrace is simple. Docker shortens your development times and accelerates your testing processes by making it easy to set up new, identical systems.
The best news is that Docker is free and runs just about anywhere. So, here’s nothing stopping you from getting started and finding out more differences for yourself!
Comments
Post a Comment