This is just a quick thought I had this morning: KV (key-value) storage solutions are excelling at item-based read/write throughput, but suck at everything that involves range queries. The column-based storage solutions might probably not have the same read/write throughput, but have a better chance at offering range queries.
I’ll probably have to check this by looking at some of the solutions included in the NOSQL reference.
Meanwhile, what do you think? Are there any other upfront ‘advantages’?
