Supporting ongoing changes with DevOps

In nature, it is not the strongest that survive, but those who respond best to change. The same goes for IT architectures, as this is where you make decisions that will greatly impact your capabilities years later. Until recently, your IT architecture was difficult to change, but now the concept of an evolutionary architecture is starting to gain momentum. The starting point here is an architecture in which change and scalability are paramount. How do you make that possible?

Developers will be well versed in DevOps, merging the roles, processes, and technology within Development (Dev) and Operations (Ops). This removes the delays and the 'battle for responsibilities' between the two so that you work more efficiently and there is less chance of failure. This approach not only makes the evolutionary architecture possible but also evolves itself again.

These are the four biggest developments in DevOps:

GitOps: the next big thing

Translated to IT, digital transformation requires a short time-to-market of applications, higher developer productivity and scalability. That is exactly what the popular open-source platform Kubernetes offers: the ability to automatically deploy, scale and manage applications in containers. If you add a version control system to capture all information, documentation and code, you get GitOps. It's basically an extension of DevOps, but GitOps has increasingly been seen as the future of DevOps since its emergence in 2017. What is the big difference and above all: why should you care? 

If your IT organization already uses DevOps for deployments, then you are most likely already working partly on GitOps. It offers the coveted single source of truth with which applications and infrastructure are synchronized. That's great for developers who work together on the same code, but it also helps the business side in achieving commercial goals. It simplifies operations, makes development faster and the infrastructure more stable. This means that as a company you can deliver a valuable customer experience that is consistent and scalable. The next question is how you ensure that you can also adapt and improve it at lightning speed.

CI/CD for fast production

Preventing and automating manual actions is of great importance for business and IT. This is where CI/CD stands out: it is the method of automating application delivery. If you want to be able to renew your application several (tens or even hundreds) times a day, you use a CI/CD framework with a set of associated tools. 

Especially when you work in large development teams, it is nice to be able to put your code into production quickly and flawless. The abbreviation CI/CD stands for continuous integration + continuous delivery or continuous deployment, depending on whether you emphasize testing or deployment. CI/CD is the backbone of modern DevOps because it provides exactly the speed you want. But speed alone is not enough. 

DevSecOps gives developers true responsibility

The most important condition for a successful digital transformation is IT security. The more containers you build and manage, the bigger your attack surface becomes. That is why it is important to include security at an early stage in the entire process. That's what SecOps (security + operations) does: it automates security tasks and guarantees that the application is of good enough quality for deployment. Part of this is monitoring your tooling. 

At a later stage, SecOps will continue as a trend in DevSecOps. This is a methodology in which not only the application/container is subjected to security aspects but also the written code, and preferably at the earliest possible stage of the development cycle. The trick is not to slow down your time-to-market with DevSecOps. By including security as a discipline in every step and from every role, it becomes a widely supported philosophy that is self-evident. With DevSecOps, every developer becomes even more responsible for security than with SecOps alone. 

Overcome cloud fear

One of the biggest underlying trends in DevOps is that IT is becoming less about hardware and more about development. This is logical given the move to the cloud, but it does not mean that you should leave the provision of infrastructure entirely to the software vendors: a fool with a tool is still a fool. Anything can go wrong in your cloud infrastructure and you need to know where to look. You do not always need many of the standard supplied bells and whistles. Without cloud management, plug & play can lead to high cloud costs. It is often better to go for a 'fit for use, fit to purpose' approach. 

Many companies also have some cold fear about the public cloud: so-called cloud fear. The DevOps mindset helps these companies with the step-by-step and structured transfer of pieces of infrastructure to the cloud. Now, public cloud changes just as quickly as the techniques surrounding DevOps, so if you really want to save those on-premise costs, you as a developer will have to be familiar with this

DevOps is a cocktail 

We mainly looked at the possibilities and opportunities of the latest DevOps trends. This gives you the ideal image of a controlled and safe development process with the shortest possible time-to-market. In practice, however, it is often more unruly. In many companies, everything is mixed up – traditional VMs and containers – and sometimes everything is completely on-premise. The well-known silos then separate the work processes, employees and technologies. 

As a methodology, DevOps offers a promising mix with which you can improve your IT processes, but then the team must of course be open to the DevOps culture. DevOps is about communication, collaboration and integration. DevOps is like a cocktail, and then something delicious has to come out of it of course.