Very nice article. I recently switched to amazon from dyn. One thing I noticed is that amazon has unique hostnames for the nameservers.

Can you comment on how Amazon's use of unique name server names impacted your results? I have a suspicion that amazon is half as fast because it requires two lookups instead of 1. ns1.dyn.org is likely to be cached, but llkadjsf.amazondns.net is unlikely to be cached.

It won't generally affect the real user test results because of a warmup/training phase in the test when those DNS records would be cached within the recursive chain.

Thanks for posting this. I thought the high end providers had much better response times than this. I was under the impression that in the US they were under 50ms for direct queries. I find it interesting you say that up to 200ms is good.

For direct synthetic tests we do see about 50ms response times in the US. For real user (recursive) tests we see lookups in the low 100ms range.