After having spent the past decade across about 12+ companies in various capacities furiously trying to make software immutable and to isolate state to dedicated services like DBs, message brokers, and networked caches, I’m not sure if it’ll go away without many companies rewriting software from the ground up effectively to support cloud native-ish assumptions like 12F but much more like avoiding leaving transactional state on-disk that can’t be recovered. There’s a TON of software out there that is written like it’s the 90s still. The market for managing legacy stateful software is matured to its finality and nobody wants to try to attempt to revolutionize it anymore is what I’m seeing, not that nobody is doing it anymore. Essentially, if one is writing new software that isn’t containerized in 2020 you’re writing the systems equivalent of COBOL.
But anyone could see many years ago that it’s easier to flatten and reinstall stuff than trying to painstakingly manage various moving pieces. We were limited by our technology unless one was using stuff like zones in BSD or Solaris.
The surprise for me has been how rapidly huge behemoth organizations like the DoD has shifted to technologies like K8S when they took so many years just to get onto VMware and some configuration management.