As you’ve probably guessed (based on the name of our company) we’re fans of Kubernetes. And for a good reason. We’ve seen companies take off pretty rapidly after implementing it properly.
Kubernetes is becoming the gold standard in the industry and there are many compelling reasons to build your platform on top of it.
But to understand when Kubernetes is a good fit it’s important to understand when it is not.
Kubernetes is not suited for MVPs
Kubernetes is a great tool that solves a lot of problems crucial to scaling your application. However, if you’re just starting out with your product using Kubernetes from the get-go is probably an overkill.
Unless you have very specific requirements that time is better spent on the product.
While you’re still validating your idea try to:
- leverage off-the-shelf solutions
- use a PaaS provider like Render or Heroku
This leads to a more general point.
Kubernetes takes time
While Kubernetes adds some useful abstractions your team needs to understand them in order to fully benefit from them.
The concepts are not hard but, as with anything, you will need to dedicate some time to learning them. Especially, if you’re doing so on your own.
Even when you have a knowledgable team building a full-blown platform from scratch is going to take some time.
You don’t need to scale
Kubernetes is wonderful when it comes to scaling your traffic. And scaling fast.
Yet, when you’re just starting out you probably don’t need to.
And even if you do, you may be fine with a simpler solution such as AWS AutoScalingGroups.
Your app is mostly monolithic
Kubernetes really shines within a microservices-based architecture.
Why?
Becase it elegantly solves problems that are inherent to it, especially around service discovery and traffic management.
But managing a monolithic app is much simpler. As such, the overhead of introducing Kubernetes to the mix might eat all time you hoped to gain with it in the first place.
That being said, Kubernetes might still be a good option if your team is already familiar with it and you’re willing to invest into it long-term.
The bottom line
So there you have it. As always in tech, it’s all about the tradeoffs.
Now that you know when not to, have a look at when to use Kubernetes.
If you have any doubts feel free to contact us – we’re happy to advise.