That said, the last two Decembers we have had trouble instantiating new instances (in US-east-1a, their default and hence most popular ___location) in December. We solved the problem just by switching to US-east-1b for some tasks.
"In order to prevent an overloading of a single availability zone when everybody tries to run their instances in us-east-1a, Amazon has added a layer of indirection so that each account’s availability zones can map to different physical data center equivalents."
