[moduspol]

I'm also excited for Spanner and Cockroach, but even at a high level, DynamoDB is essentially requiring you to design how you're storing your data based on how you will access it, how it will be sharded, and how it will be pruned. That makes a lot of use cases tricky, but the end result is that if you design right, it will scale linearly quite well.

It feels like using a relational model (with SQL) is fundamentally different in that you're designing how you're storing your data based on how it relates to other data. It's certainly a whole lot easier to design for, but it makes me wonder if it is possible to close that gap in terms of "easy to design / SQL / vertically scalable" and "serverless / cheap / horizontally scalable".