Just wanted to chime in and hard agree on this. I remember the world where people were trying to build things like kubernetes before it existed. There was a period where enough people had left google post ipo to realize that they needed something like borg but nothing really existed like that outside of elgoog.
Twitter had Aurora, then elsewhere mesos and mesosphere popped up and it seemed like there were going to be about 100 different frameworks until Kubernetes dropped and basically ate the industry alive. K8s feels pretty stable from where I’m sitting.
Not to mention all the various in-house systems in the 2010-era. I worked at a small place back in 2013 that had built their own container platform about a year before Docker was released.
It was more like workqueue than borg. But it worked well enough. Eventually we got tired of maintaining our own snowflake scheduler and switched to Kubernetes.
But I'm also old enough to remember pre-borg systems used in HPC. Maui, TORQUE, etc. I wasn't surprised batch scheduling finally got used for service deployment.
Even in the post docker world people were trying to build container platforms on docker that resembled other things but had their own issues. I definitely had to maintain one of those for three years.
Kubernetes is kind of stable except for the relentless deprecation and introduction of features. I.e. ingress going from v1beta1 to v1 and dropping any support for the v1beta1 defs, service accounts no longer having tokens associated with it, etc.
It really makes for a constant churn where there's a good change something breaks if you update kubernetes. Either that or you pick a version and stick with it for 2 years while it's supported and then jump to the new one and fix all the breakage.
Twitter had Aurora, then elsewhere mesos and mesosphere popped up and it seemed like there were going to be about 100 different frameworks until Kubernetes dropped and basically ate the industry alive. K8s feels pretty stable from where I’m sitting.