We have built our infrastructure to be able to handle lots of requests for lots of different resources and data. However, all subscription plans dictates a maximum number of requests per second (burst rate) and per minute, and we urge all users to follow sound development practices and not make unnecessary requests. For large user base apps that consume the API, we recommend using a reverse proxy to cache and handle authentication towards the API. Especially since this will cache your apps specific request/data patterns as efficiently as possible and speed up your app.
We always reserve the right to block clients with an unreasonable amounts of requests or ones that seem to have been compromised or are otherwise erroneous.
We run live analytics on our API and the requests that go in and out of it. If we notice that a client has reached or exceeded its limit, or is behaving abnormally, we contact the client to figure out the situation. It is usually only a matter of bug fixing or tweaking required to fix such errors. However, if the problems persist, we may require the client to upgrade their subscription plan to continue using the API.
Updated about a year ago
Lets have quick look at the biggest changes since v1