In the tech industry, the term « DevOps » is becoming increasingly crucial for modern software development and distribution. As an active or aspiring engineer, you’ve almost certainly encountered this term at least once, if not daily. DevOps is surrounded by a lot of hype and debate. Which approach is best ? What skills and tools are important to study ? Do skills and tools still matter ?
Capmad analyzes the modern DevOps landscape and how you can develop your skills to become a DevOps engineer yourself.
DevOps : Brief overview of the profession
The term itself is a portmanteau of « development » and « operations ».
The initial goal of the DevOps movement was to address the growing frustration caused by operational and technical silos that often separated development teams from system administrators.
With DevOps, developers and administrators work together with a shared understanding and culture focused on continuous improvement and faster software delivery. It’s not necessarily about choosing the latest and most advanced tools and programming languages.
A famous DevOps axiom is People > Process > Tools. While there are nuances to this model, at a basic level, it provides a clear understanding of what is truly essential to achieving one of DevOps’ main goals: increasing deployment speed.
Why learn DevOps ?
DevOps ranked among the top tech skills to acquire in 2023 by Pluralsight based on market research and learner trends. Here are some figures that show why it’s worth adding this skill to your repertoire:
- 59.3% of companies reported having a DevOps function.
- DevOps was listed as an « essential » skill by Gartner in 2023.
- Over the past three years, DevOps has been one of the most searched terms on Pluralsight and A Cloud Guru.
- DevOps Foundations was the ninth most popular skill tested by IQ on Pluralsight.
If you’re looking for internal DevOps opportunities, now is also the best time to upskill: 8 out of 10 tech companies plan to invest in developing their technical skills in 2024. In other words, it’s the perfect time to work on your upskilling pitch.
How to learn DevOps : 8 crucial areas to improve
Linux operating system administration
Linux is the operating system of choice powering the most modern internet infrastructure. Most mobile apps, SaaS/PaaS/IaaS platforms, and websites you interact with daily are powered by Linux. Mastering Linux is essential if you want to work in software distribution management in these environments.
Basic networking
The shift from on-premises networks to racks of switches and routers has reduced the demand for specialized and vendor-specific network certifications. However, general networking knowledge remains highly valuable, and it’s important to understand how networks behave and interact within a cloud platform.
Interpreted language (Python or TS)
Unfortunately, this is often where many people falter. If you are a software developer aspiring to start a DevOps career, you are already ahead in one of the most challenging technical skills to acquire. Fortunately, knowing a language like Python or TypeScript well enough to succeed in DevOps does not require a full computer science education. Self-learning and practice are more than sufficient to gain a solid foundation.
Cloud Services/Platforms
Every major cloud platform now offers a free tier of services that clients can use to test and learn without incurring usage fees. You can also find additional credits offered by various providers and platforms. Start by creating a basic application infrastructure and, as you become familiar with it, gradually increase the complexity.
Version control
Learning to manage versions is another skill dependent on your experience. The tool choice is relatively simple: go for Git. It is almost standard at this point, and SVN and Mercurial are likely to be found only in legacy environments. The two largest VCS platforms, GitLab and GitHub, are both based on Git; it’s the de facto choice. For system administrators, you will need to practice Git workflows that reflect what you would experience in an environment where you need to collaborate with developers and other DevOps engineers.
Monitoring/Observability
Monitoring and observability are essential elements of the DevOps approach to software development and delivery. Adhering to a principle of continuous improvement means having a precise and comprehensive understanding of application performance and infrastructure status. If deploying a feature causes application slowdowns for end users, it’s crucial to quickly identify and resolve the root cause.
CI/CD
One of the fundamental pillars of modern software distribution infrastructure is the continuous integration/continuous delivery (CI/CD) pipeline. This tool serves as a « base » for automating most processes related to testing and deploying applications. CI/CD pipelines have enabled a massive increase in deployment speed, allowing for multiple deployments per day. You can also use DORA measurement tools to track modification lead times and deployment frequency.
Infrastructure as Code (IaC)
You’ve probably heard about Infrastructure as Code (IaC) and DevOps. The two seem almost inseparable. While you can certainly undertake a DevOps initiative without it, IaC unlocks features that are nearly indispensable if you want to manage and deploy software automatically in a modern cloud environment. IaC allows engineers to define infrastructure configuration in code, rather than manually clicking through a user interface to deploy resources.
Conclusion : Do more of what you already know !
If you’re already in the tech sector, diving into DevOps doesn’t necessarily mean starting over. Do more of what you already know, look for opportunities to develop your skills, and spend time learning new technical skills.
