Google Cloud IoT vs. K3s Lightweight Kubernetes

I consider myself very techy and found Google Cloud IoT platform very challenging to manage. The lack of tutorials and discussions to understand how each section works is very challenging. I specifically made a connection after several hours to Google Nest to third-party integration, Home Assistant. Shortly after Google Cloud upgraded to a new version breaking the connection. This was extremely frustrating. No service should take several hours to figure out, in my opinion, if it does, the platform is doing a poor job of making it easy. I'm personally very discouraged any time I ever have to use this platform. It's very hard to find answers.

Incentivized

K3s Lightweight Kubernetes is well suited for local development. K3s Lightweight Kubernetes is not well suited when you need backward support. In the case that you want to test an application that is still using deprecated APIs or CRDs which are supported on popular managed Cloud Kubernetes Distributions, because K3s Lightweight Kubernetes is trying to be a lightweight application and dropping early support for APIs/CRDs marked for deprication K3s Lightweight Kubernetes will not be a good solution. If you are trying to test Kubernetes etc, functionality, K3s Lightweight Kubernetes is not a good case since it's using Dqlite or SQLite.

Incentivized

• Integration with different brands of microcontrollers including the one currently used by Espresiff.
• The platform is very robust and secure.

Incentivized

• Virtual clusters
• IoT devices
• Local development
• Raspberry Pi

Incentivized

• Not beginner friendly.
• Needs more tutorials and walk throughs on how to get started.
• Very complicated and unless you know what you are doing you will be lost.
• Lack of material found on how to do each section / services.

Incentivized

• Option to deploy a self managed k3s in the cloud
• Support to deploy it with kubeadm
• Integration with prometheus & grafana

Incentivized

Although comparisons are hateful, even more so when we are talking about leading brands where the quality of their services are indisputable, the general environment of Google was more familiar to me since I use, for example, Google Firebase on a daily basis, where part of the concepts are similar, without Without a doubt, AWS services are excellent, but it was easier for me to go through the functions of Google Cloud IoT.

Incentivized

We have mostly used kind, minikube, and Docker Desktop Kubernetes for local development. Only kind has lower resource requirements, but K3s Lightweight Kubernetes seems to be even lighter. It is a big benefit to be able to deploy a K3s Lightweight Kubernetes cluster in seconds locally in our workstation, there is no cost, and it makes it easy to redeploy from scratch no matter how many times we want. The removal of deprecated APIs or CRDs also makes it more secure since it reduces unnecessarily exposed attack vectors.

Incentivized

• The amount of hours to get things integrated is negative.
• The amount of hours researching how to get devices integrated is negative.
• Overall the amount of time and effort getting things working has been a negative experience.

Incentivized

• Engineers have been able to run more microservices locally with K3s Lightweight Kubernetes instead of Minikube, which is more resources hungry

Incentivized