Of course. What I meant is that if you have a dataset for which you can give an answer by loading it all into memory on a single machine -- it's not a big problem. That isn't to say that every problem which is better solved using a distributed system is a big problem.

This times 1000. Most developers don't realize that we're just starting to circle around again (this is a known, common pattern) and rediscover the reason for complex distributed systems from our past. Now we'll just do it with shinier tools.

As problem spaces expand beyond our current hardwares ability, we start distributing it until a new wave of equipment becomes available and we start labeling our predecessors idiots for having ever built something so complex.

Yes, and while you can solve many problems with parallelism, that doesn't mean you should.