We do not plan on using 180 requests a second. This is just an exercise to understand how your limits our built so that we can build our API batch processor to work efficiently but always below the limit. If you are interested:
- We will never come even remotely close of hitting the “light” rate limit of 80 calls/s
- Until webhook retries are re-enabled, we will be making a good number of “medium” calls. It is highly unlikely we would ever have close to 60 calls/s (and we will have an internal batch processor that will keep us from exceeding that rate) but we try to plan for peak usage, so knowing that we have the full 60 medium calls per second available to us even if there are light or heavy calls that also happen that second is helpful.
- We will have much fewer heavy calls and no resource-intensive calls.
Webhooks are a great way to build an integration, but our clients rely on our features to work 100% of the time. Without retries enabled we have to build API requests into our application as a fail-safe. We will keep those under the rate limits, but there will be a good number of calls.