Judging by the buzz it has created, this last year has been the year of Kubernetes. It is everywhere, and as usual with such popular subjects, there is a huge stash of content about it: tutorials, blog posts, talks by the dozen, everybody seems to be speaking about Kubernetes… So everything is good in All is well in the best of worlds, right?
Well, to be sincere, I wouldn’t say so… Most of those tutorial, posts and talks stop just after they have guided you to do the “Hello Kube”, when you have a handful of applications deployed on a local Minikube on your laptop. But that is not the end, it’s more of the true beginning of the path…
In this talk I’m going to speak about the gap between Minikube and a production-ready Kubernetes infrastructure, about the dozens of “small details” that become show stoppers the first time you try to deploy your production Kubernetes, about the usefulness of managed Kubernetes solutions, about vendor locking, about multicloud and Kubernetes federation. In brief, about what to do after you have deployed on your Minikube…
In the second part of the presentation, I will address a different public, the Montpellier JUG (Java User Group) attendees.
What does a Java developer can do in a world where containers and Kubernetes seems to have replaced traditional Java and JEE stacks? How can the JVM compete with the blazing fast and light containers from other stacks? Is Java doomed to become the next COBOL?
Fear no more, the Java community has well understood this brave new world, and solutions are available.
In this talk I will talk about GraalVM, a blazing fast, polyglot, lightweight, compiling to native JVM giving you the capacity to compile your Java applications to small native binaries easy and painless to deploy in containers and able to compete with any other Cloud Native stack.
And then I will talk of Quarkus, and how you can combine it with GraalVM to create powerful Java applications, in a modern way.