What cloud to learn in 2022?

If you’re interested in shifting your career into DevOps, you probably know by now how important the cloud is. But there are so many options out there. What cloud should you learn in 2022?

The short answer - learn AWS. The long answer - scroll down and let me explain why 👇👇👇

Everything in the world has a purpose, especially decisions to try new things. Learning cloud (or anything for that matter) is not an exception. It is important that you know not only what you want to learn, but why you want to learn it. How will this serve your personal purpose?

The ultimate question of life, the universe and everything

So, before you proceed with this article, ask yourself a question: “Why do I want to learn cloud?”

Most likely, your reason to learn cloud (or any other technology) is that you want to get new skills that will make you more attractive to employers and finally give you a better high-paying job. These are all valid reasons, but they don’t help you narrow down how you’ll achieve these goals and what cloud to learn. I encourage you to dig deeper.

Try this approach to the question, in a slightly different form: “What cloud should I learn to maximize my success on a job market?”

What cloud will maximize success on a job market?

I wanted to answer this question precisely, based on real data.  I analyzed several dozens of job postings to see what cloud technologies they list in their job description. As a baseline keyword for my search, I focused on the title of “DevOps engineer” (and before you come for me, yes, I recognize this isn’t the most precise term because DevOps is a methodology). I limited my search to the New York City area. And here is what I found.

More than 50% percent of job postings require AWS. This means that if you know AWS, your chances of being a fit for a job are two times greater than if you know GCP or Azure.

This does not end with cloud though

This analysis clearly shows that AWS is the cloud to go with if you’re looking to maximize your potential for success on the DevOps job market. It doesn’t end with the cloud though; we can use this analytical approach to make informed decisions about any tool, technology, or framework.

For example, we can decide which operating system is better to learn.

67% of hiring managers have Linux as a requirement in their job postings – it’s more than two times more popular than Windows. This means Linux should be your first choice of the operating system for matching requirements for the maximum possible job opportunities on the market.

Kubernetes again?!

How about Kubernetes? You’ve probably heard of it growing in popularity. Is Kubernetes really that hot? 🔥🔥🔥

The data tells us that Kubernetes is a requirement in more than one out of every two job postings. With a 50/50 chance of Kubernetes being expected in a DevOps job, you certainly gain an edge if you decide to add it to your resume.

What about Docker? Docker was the first container runtime that was supported by Kubernetes. Even though Kubernetes has recently deprecated Docker as a container runtime, they still come hand-in-hand with each other in job postings, so it’s helpful to know both. 🐳🐳🐳

Should a DevOps engineer know how to code?

Does a DevOps engineer need to know programming? A typical hiring manager will likely say yes. Some programming language is mentioned as a requirement in 67% of jobs.

That can sound overwhelming because learning programming from scratch is challenging and time-consuming. When it comes down to it though, according to this data, you don’t have to worry too much. In most cases, programming is not a primary requirement, but rather nice-to-have.

Ok, we have identified that we need to know programming as well, but which language to choose? Among programming languages, Python is the absolute leader and Bash is on the second place.

What automation tools to learn?

Jenkins has crashed everybody on a CiCd front, along with GitHub as a heavily preferred version control tool.

When it comes to Infrastructure as Code, Terraform (which is my favorite IaC tool) is an absolute winner in the industry as a whole, with a small fraction of postings that cite CloudFormation and Azure ARM. Even though Pulumi had a good start when it was first released, it hasn’t reached global popularity—it was only a requirement in one position.

In the Configuration Management field, the biggest player is Ansible. It’s required in almost 50% of cases. The second place is shared by Chef and Puppet with a small pinch of SaltStack. 🧂🧂🧂

What about degrees and certifications?

A lot of people who are pursuing DevOps as a new career don’t always have a higher degree. 🎓🎓Often, this can cause anxiety and make people feel disadvantaged for a job in the field. But when we ask “is a degree a requirement?” in this data, the answer is - not really.

Only 23% of jobs require a degree, with an additional 9% that would like to see a candidate with a degree but are also ready to accept experienced candidates that don’t have one. The remaining majority—66% of jobs—don’t mention anything about a degree requirement at all. This is an industry you can truly enter into at any point in your professional life even without “traditional” education or degrees.

Finally, when it comes to judging how qualified a candidate is for a job, are companies more partial to certifications or experience? How valuable are certifications from a hiring manager’s point of view? Unfortunately, only 4% of jobs mentioned something about Certifications.

Where certification was mentioned, it was only seen as a nice-to-have supplemental feature, but not a requirement by any means. That said, there’s certainly no downside to being certified. Even if not a job requirement, it can help you stand out from the pack.

Let's summarize

Certainly, data can show us so much about trends and job requirements in the DevOps field. One of the reasons I love this exercise is because it objectively analyzes job postings to give you a visual breakdown of whatever you’re studying. Even more, it’s replicable—you can conduct a similar study yourself, focused specifically on your location and keywords.

As a guide, here’s a general process for conducting similar research on your own:  

  1. Define your target job position and location. For example, in this study it was “DevOps engineer” in New York City.
  2. Search a job database using your keyword and count how many times a technology (e.g., cloud, Kubernetes, Windows, certificate, etc.) is mentioned in the dataset of job postings
  3. Compare your results. It can be helpful to create graphs and visuals to quantify the comparisons.
  4. Make an informed decision about next steps. Use this data to guide you as you move forward in your DevOps career. Figure out what to prioritize and focus on. And then…start learning!

And of course, any career journey is faster and more engaging with an experienced friend who can help out and give personalized advice. Luckily, we offer that here at Your DevOps Mentor—where our recommendations are always research-based just like this one.

Apply for a mentorship here: https://yourdevopsmentor.com/apply/

Connect with me on LinkedIn: https://www.linkedin.com/in/vladimir-mukhin-devops/

The success story of my client: https://yourdevopsmentor.com/blog/from-biologist-to-devops-engineer/

Originally published Jan 31, 2022 3:08:25 AM (updated January 31 2022 )

Leave a Reply

Your email address will not be published.