Rate Limits

The Connected Sensors API enforces a rate limit of 4 requests per second per token. Exceeding this limit will result in 429 Too Many Requests responses. This applies to all API operations — sending device data, reading variables, querying history, and any other endpoint.

Understanding the Limit

This limit applies regardless of the type of operation — both data ingestion (POST/PATCH) and data retrieval (GET) count toward the same 4 req/s ceiling.

Best Practices

Batch your data

Rather than sending one request per variable, group multiple variables into a single request body. Most endpoints accept arrays of values, letting you send an entire device's data update in one call instead of several.

Throttle polling requests

If your integration queries the API on a recurring schedule, ensure requests are spaced at least 250 ms apart. Running multiple concurrent queries against the same token compounds the rate quickly.

Coordinate shared tokens

If multiple processes or services share a single API token, coordinate them so their combined request rate stays below 4 req/s. Consider using separate tokens per service where possible.

Use bulk endpoints

Where the API offers endpoints that accept arrays or bulk payloads, prefer those over making individual calls in a loop.

Handling Rate Limit Errors

When the limit is exceeded, the API returns:

HTTP 429 Too Many Requests

If you encounter this in your integration, implement exponential backoff — wait a short interval before retrying, and increase the wait time with each successive failure. This prevents a burst of retries from compounding the problem.

Retry after: 1s → 2s → 4s → 8s ...