How many requests can scanii process concurrently?
In theory it’s unlimited since our systems will keep auto scaling to meet your throughput needs but, in practice, there are a couple of things to keep in mind:
- Certain resources are protected from abuse via rate limiting logic exposed to you via github-style X-RateLimit-Remaining and X-RateLimit-Limit headers telling you the number of tokens (api calls) you can use during a window of time.
- If you blast us with a rapid expansion in API calls, let’s say going from 1,000/s to 100,000/s you might experience a brief amount of 500 level HTTP errors as we harvest more cloud capacity behind the scenes to meet the higher throughput needs. This process is fast but it isn’t instantaneous so we advise clients to setup some sensible retry logic around calls to our API, for example, in the event of a 500 level error, wait a couple of seconds and try again.