Hi! I'm working on an application which does a few consecutive MusicBrainz indexed queries (I'm using the python-musicbrainzngs bindings) but in a pretty time-critical way for which we don't want to keep a user hanging around. Right now for dev purposes, I am just connecting to the webservice on the main site or one I have running locally in the pre-packaged .ova virtual machine image, running inside VirtualBox and at best I have limit=100 queries taking a few hundred milliseconds. Can anyone speak to what my expectation of turnaround time might be if I had everything configured on one of the more powerful EC2 instances, with my application also running on AWS? While I plan to put a caching proxy between my app and the server to take care of repeat requests, I am still interested in requests that do go through to the backend to run as quickly as possible. Apologies if this is something of a 'how long is a piece of string' question - I will follow up if I have inadvertently omitted a crucial piece of information. Thanks!