All of these posts never seem to talk about the operational overheard they're saving by using the cloud. 500k/year is a couple of engineers. And IME, at scale, Elasticsearch is a huge pain to self manage.
Self hosting has become very easy these days. Elasticsearch, specifically was a pain to self manage years ago. With modern versions, it has become incredibly stable, defaults have made more sense, and you now have modern tools like Kubernetes to make deploying/scaling easier too. The work required to self-host has improved in the past decade, and ppl who "default to the cloud" ignore this.
I'm saying this as someone who manages a 50 TB cluster for a startup (which I know isn't a HUGE amount, but definitely not small either)
I'm saying this as someone who manages a 50 TB cluster for a startup (which I know isn't a HUGE amount, but definitely not small either)