Why do I hear Docker is no longer in demand?

I’ve heard recently that Docker skills aren’t in demand, rather the demand is on stuff like Kubernetes? I personally have had a tough time working with Docker, and my head explodes the momment I try to figure out why and how I should use something like Kubernetes.

Anyone with experience with either of these common DevOps tools?

2 Likes

Heard from whom/where? Whatever, it’s hogwash, containerization is bigger than ever, and Docker is seen as synonymous with containers (there are others, rarely used). Kubernetes (aka “k8s”) is a platform for orchestrating and deploying containers, much like Docker Swarm. You’d use Kubernetes when you need multiple containers working together (aka “orchestration”) like, let’s say a database container, a webserver container, and a redis container. Kubernetes lets you define your different containers as one logical group, have them share a filesystem, connect to each other over a private virtual network, and start/stop them all with one command. If one of them goes down, you can automatically restart it, and your web traffic gets a spike, you can have it spin up more webservers automatically.

My shop is all cloud-based and going 100% containerized, and we don’t have any plans to use Kubernetes. It’s really made to manage single organizations, whereas we have several clients with no relation to each other whatsoever. Doesn’t help that k8s on AWS isn’t free, whereas ECS is.

3 Likes

I’ve heard (can’t find any references anymore sorry) containerization is bigger than ever, but just knowing Docker isn’t good enough anymore. As you mentioned, Docker Swarm, and Kubernetes are orchestration software that uses Docker.

So yea you need to know Docker, but its only a tiny piece of the puzzle, and thus saying “I know Docker” is only part of what your expected to know.

I also don’t use Kubernetes due to its complexity and find it hard to apply to any day to day work.

I see on full-stack job offers that they like to put “Kubernetes” keyword now. I am guessing that mostly they are just listing their full tech stack they are using.

I meet quite many devops at tech events who never used Kubenetes… So personally I plan to learn very basic stuff with it, just enough to start a container.

One project that looks very cool to learn kubernetes, is to combine multiple raspberry pi to reproduce a mini data center, definitely on my todo list for later.

Chiming in here!

I can say with good certainty that docker is not going away (or at least containerization for that matter).

Depending on the maturity and the scale of the company you are working for, docker becomes a necessity rather than a nice-to-have. It adds to reliability in builds (through both development and especially through production), and it supports testing extensively. This is because Docker provides reproducible environments, building your system the exact same way each time you run a container from your specified image.

This gets exacerbated in environments that require distribution across multiple areas, say for example clients in North America and Asia.

That’s where Kubernetes comes in, as @chuckadams mentions it allows for scale, redeployment, and orchestration.

You can have your system or platform distributed across differing regions with redundancies that either scale with use or redeploy upon error, each rebuilding under the instructions of their underlying containerization (docker, in most instances).


I have used both extensively where I work, and I can say that it is definitely worth your time to invest in learning the basics of docker. It helps with open source projects; it will help your prospects at newer companies and bigger companies that are keeping up with technological paradigms.

A small bonus of learning docker as well: a lot of the concepts learned from a basics docker course will carry over to anything related to an orchestration implementation as well. Kubernetes and docker share a lot of the same core concepts and docker is a major component of the kubernetes stack.

Finally knowledge in docker will help you synthesize knowledge around both items like gitlab runner, and github actions (among other pipelines) both of which are effectively extensions of the docker environment.

:slight_smile: