5 predictions for Kubernetes in 2020

And, a review of the most popular Kubernetes articles of 2019.

How do you track the trajectory of a popular project like Kubernetes? How do you know how far it’s gone? If you are contributing to the project or joining a special Interest group (SIG), you may be aware of its trajectory in a small way, but if your full-time job does not involve contributing to Kubernetes, you may need a bit of a glimpse into the future to help you understand. For a fast-moving project like Kubernetes, the end of the year is a good time to reflect on the past year and look ahead to the new one.

This year, Kubernetes has made great progress. In addition to looking at source code, documentation, and meeting notes, you can also browse blogs. To dig deeper, I found Kubernetes’ top 10 articles at Opensource.com. These articles give us an idea of what topics developers like to read and write about. Let’s get started!

Why do data scientists love Kubernetes
Kubernetes confidential information and an introduction to ConfigMap
How to run PostgreSQL on Kubernetes
Why Kubernetes is a dump truck
Safely scan your DevOps process
Deploy InfluxDB and Grafana on Kubernetes to collect Twitter statistics
Extend PostgreSQL with the Kubernetes operator
Use Kubernetes controller to reduce system administrator workload
Bring Kubernetes to the bare metal edge for calculation
Why don’t you worry about Kubernetes

First, I should point out that five of these articles are about extensions to Kubernetes workloads and what scenarios they can run in. These workloads cover data science, PostgreSQL, InfluxDB, Grafana (not just monitoring the cluster itself), and edge computing. Historically, both Kubernetes and containers were run on virtual machines, especially when running on cloud-provided infrastructure. Despite the interest in Kubernetes, this shows that end users are desperate to install Kubernetes on bare-metal machines (see Running Kubernetes on bare-metal machines with OpenShift).

Secondly, there are many developers who want to learn about operations and Kubernetes best practices. From the Kubernetes operator to the Kubernetes controller, from confidential information to ConfigMaps, developers and operations are looking to find best practices that simplify deployment and administration. We often obsess over how to change a configuration file or how someone else will configure it, instead of thinking about how that configuration makes the application deployment work (not how to install it, not how to run Kubernetes).

Finally, people seem genuinely interested in the introductory tutorials. In fact, the amount of information needed to build Kubernetes was so overwhelming that it discouraged people and led them down the wrong path. Several of the popular articles explain why you need to know how to run your application with Kubernetes, not just install it. Just like best practices articles, people often don’t go back and analyze where they should have spent their time in getting started. My philosophy has always been to invest limited time and money in how to use a technology, not how to build it.