Nowadays, Kubernetes has become one of the most popular platforms for many workloads; from databases, to data streaming systems, to microservices-based applications. However, managing these workloads using only the Kubernetes native resources is not easy. Especially when it comes to day 2 operations like upgrades and scaling. Helm charts can help to a certain extent, but they don’t solve all the potential problems.
What you need is an operator, not a human one, looking after your Kubernetes workloads 365/24/7. In the end, this is how the internal mechanics of a Kubernetes cluster works, so why not use the same approach for your own applications?
During this session we’ll explore what the “operator pattern” is and how a software-based operator, with the necessary domain-specific knowledge, can take care of your Kubernetes workloads; helping with installation, upgrades, certificate management, and reducing the need for human intervention. We’ll also demonstrate a real operator in practice - the open-source Java-based operator Strimzi which manages Apache Kafka clusters in a cloud-native way.