Not to be "old man yells at sky" but a lot of these new cloud-based/cloud-focused architectures seem to be geared toward highly specialized needs that 99.9% of businesses aren't going to need. However they do one important thing - they over-use resources that line the pockets of MS, Amazon, Google, Data Bricks, etc. A Data Lakehouse is fine but what benefit does it give you over a much more simple solution of ETL/ELTing the data in batches (weekly, daily, hourly, etc) and letting it sit in some kind of DB.

They say the Data Lakehouse needs all this metadata storage, API access layers, etc. Seems like an overly complex system for anything but large real-time systems that need to replicate a DB but due to data volume and throughput, are unable to. Perhaps you also aren't just driving traditional reporting (dashboards, etc).

I'm happy to use this new technology to make more money for myself as a specialist, and effectively be in on the scam, but from an optimal solution pov they suck.