1) we will release on premise version to address that (yes, you can deploy more machines to do HA and load balancing) 2) AWS allows us to get new IP addresses

1) we will release on premise version to address that (yes, you can deploy more machines to do HA and load balancing) 2) AWS allows us to get new IP addresses quickly and APIs will ban rather the auth key than the IP address no? There are huge networks behind NAT like mobile and companies so providers has to count with that. 3) We enforce single point of entry to the Traffic Monitor through the generated unique URL and key we generated for you. The monitors are jailed on several levels, so they can't access outside resources. 4) We are having optional SSL now but enforcing it in next days. 5) We don't want to enforce any limits other than abusing the service. Do you have any numbers to share? We were thinking about 10req/s should be enough for everyone. 6) It should not double the response time because we are not slow as the API you are asking. Of course it depends where is your client and how many request we will be having, but we have plan for spreading the load and moving monitors around. Also 1)(on premise) is in the works which should have almost 0 latency. Thanks! If you are interested write us a mail and we can chat how to improve things.