Your usage of the API is covered by our terms of service. This specifically includes the requirement that "You agree not to engage in any of the following prohibited activities: ... taking any action that imposes, or may impose at our sole discretion an unreasonable or disproportionately large load on our infrastructure". Currently we interpret this to mean no more than 120 updates (POST request) per minute, 120 single item fetches (GET specific entity) per minute, and no more than 300 collection requests or reports per hour.
Applications that exceed these rate limits may receive HTTP 429 Too Many Requests errors when they hit the rate limit. Once hitting the limit, requests for the API will fail with that error code until the next time period (minute or hour depending on the specific API endpoint) starts. API endpoints that have automatic enforcement will return the following HTTP headers to assist with staying under the rate limits:
- X-RateLimit-Limit: The total number of requests they can make in a time period.
- X-RateLimit-Remaining: The number of requests they can still make before the reset time.
- X-RateLimit-Reset: Unix timestamp of when the rate limit will be restored.
Please contact us to discuss your specific application if you expect to be close to these request rates so we can discuss your application and how it can be optimized to run efficiently on our infrastructure.