Rate limiting

DocuMine Documentation

Rate limiting

Please note, that the API is designed to ensure optimal performance and fair usage by implementing rate limiting on API requests.

It uses a token bucket algorithm to manage API request rates. This approach helps control the amount of data a user can request from the API over a given time period.

The average bandwidth of allowed requests per second can be calculated using the formula:

Average requests per second

Using the given parameters, we get an allowed average 10 requests per second.

The token bucket algorithm allows for occasional spikes in request volume, enabling developers to temporarily exceed the average request rate without triggering rate-limiting errors. This flexibility accommodates traffic bursts and ensures uninterrupted service delivery. Additionally, services can be scaled to manage a higher volume of requests when needed.

When the rate limit is exceeded, the API will return an HTTP 429 Too Many Requests. status code. This error indicates that the request cannot be processed because too many requests were sent within a short period.

Best practices for handling 429 errors

Exponential backoff:
Implement an exponential backoff strategy to retry requests at increasingly longer intervals until they succeed.
Rate limit headers:
Pay attention to the rate limit headers returned by the API, typically including X‑RateLimit‑Limit, X‑RateLimit‑Remaining, and X‑RateLimit‑Reset. These headers provide details on the maximum allowed requests, the remaining requests available, and the time until the rate limit resets.
User notification:
Notify users of your application when rate limits are reached and requests are delayed to enhance the user experience.

In this section:

Was this helpful?

DocuMine Documentation

Rate limiting

Search results