It's significantly more difficult to scale a traditional relational database (although not impossible!), than to scale the web/app layer that sits in front of i

It's significantly more difficult to scale a traditional relational database (although not impossible!), than to scale the web/app layer that sits in front of it. Snapshot + clone + some kind of sync middleware (like pgpool for postgres) can probably get you 80-90% of the way there. Rearchitecting so that your db server is not the bottleneck should help there as well. Maybe you need to have a master/slave setup, and on huge load, flip the slave instance over to be a instance type with quadruple the RAM and CPUs for a few hours, then back to a single-core, low-memory instance to keep the data-sycn flowing. There's a million ways to skin this cat. If your database itself is the bottleneck, then, yeah, on the fly flexibility might be difficult to achieve. In his case, a relational database probably isn't the bottleneck at all, and scaling out caches, web front ends, etc. is all fairly straight forward. There are huge numbers of folks taking advantage of this kind of flexibility. Hell, Amazon has a whole API you can integrate with that handles it for you (even has $ references, so you don't accidentally spend yourself bankrupt because of a TC story).