This message was deleted.

I think that Kubernetes can be too complicated in many cases. My opinion on this topic (from PlatformCon):

Just watched it, such a great point of view!

Its definitely a system that requires an expert / operator. If you want to take advantage of it but not run it, AWS Fargate, GKE AutoPilot are some options. GKE, AWS EKS, AKS required experts to run. Use Heroku, GCP Cloud Run etc until you cant anymore We've been running Kubernetes for 5 years, 8 clusters and growing. ~150 engineers including 20-30 DevOps engineers

Hi @Andre Marcelo-Tanner, thanks for your recording. This is actually a question that's been on my mind for some time now. I don't have a lot of AWS experience, but a lot of GCP and I know AWS Fargate is similar to GCP's cloud run in terms of intended functionality. I did understand your point on 'vendor lock-in', referring to the complexity of K8s. I agree with you on that. So far, one of the main reasons I think K8s is preferred over Fargate or GCP's cloud run may be the ability to operate workloads across different cloud platforms and even on-premise, therefore avoiding lock-in with one cloud vendor. But all that is still too much effort, especially for smaller organizations. I've also heard some explain their choice of K8s arising from ease to integrate with other companies, probably because they would have more flexibility since they are more in control of the entire architecture as opposed to using a managed service.