Rate Limit
To protect APIs, the system enforces rate limits to avoid overload when one API is called too frequently in a short time.
If you send too many requests at once, the API returns HTTP 429 Too Many Requests.
You should manage API usage carefully to avoid exceeding limits and impacting service stability.
API limiters
We apply multiple limiters, including request-rate limiter and concurrency limiter.
We may adjust limits to prevent abuse. If you need higher limits for high-traffic workloads, contact SimplifyTrip Support. For large limit increases, contact us at least 6 weeks in advance.
Rate limiter
A Request ID is a combination of [IP Address], [HTTP Method], [Mode], and [Api Key]. Request counting is based on Request ID.
Rate limiting uses the Leaky Bucket algorithm with a 3-second window. Limits reset every 3 seconds.
Method limits:
| Method | Limit |
|---|---|
| GET | 2000 requests / 3 seconds |
| POST | 100 requests / 3 seconds |
| PUT | 100 requests / 3 seconds |
| DELETE | 100 requests / 3 seconds |
| Other | 50 requests / 3 seconds |
Concurrency limiter
Concurrency limiter restricts how many requests can be active at the same time. Concurrency-limit violations often indicate long-running or resource-intensive requests.
Note This rate-limit mechanism does not apply to inter-service API calls inside SimplifyTrip internal systems.
Common causes and mitigation
-
Bursts of high-volume requests can trigger limits. Usually this happens during analytics or migration workloads. Control client-side request rate.
-
Many long-lived requests can trigger concurrency limits. Endpoints with expensive operations may take longer. Monitor request durations and timeouts.
-
Sudden traffic spikes (for example flash sales) can trigger limits. If you expect a spike, contact SimplifyTrip Support beforehand.
Handling rate limits
-
Detect HTTP
429and implement retries with exponential backoff. -
Add jitter to backoff intervals to avoid thundering herd effects.
-
Apply global traffic shaping on your side, not just per-request optimization.
-
Implement client-side rate limiting that tracks request volume over time.
-
Cache API responses where appropriate to reduce unnecessary calls.
Load testing
Many teams run load tests using Test Mode. We generally do not recommend this as a direct proxy for production because Test Mode limits and behavior can differ from live environment.
A better approach is to build configurable mocks for SimplifyTrip API in your load-testing setup and simulate realistic latency profiles based on live measurements.