ClusterMonkey has an interesting case study that examines what kind of cluster your could build today for $2,500 versus what you could do (and what they did) 18 months ago.
It’s a little long, but an interesting study in the design tradeoffs that lead to a balanced system at any scale.
But more importantly, I’ve learned that in 18 months technology has really changed. For $2,500 we have gone from eight 32-bit cores at 1.7 GHz to eight 64-bit cores at 2.4 GHz, from 2.3GB of total memory to 8 GB of total memory, from one GigE network to two, and we doubled our disk space.
In case you’re wondering, I’d rather use it to bribe an Apple official to give me an early iPhone.