APIM policy rate-limit-by-key with a dynamic numbers of calls
Hi All, We have an application that is running on kubernetes with auto-scaling enabled. This app is an API that we expose to multiple clients and it is fronted by APIM. Currently, we're facing an issue where we see our system falling apart if the API call rates/second double or triple. For example, if we were serving 100req/sec and then suddenly the requests spikes up to 300req/sec the system falls apart because there's no enough resources and the auto-scaling need at least 30 sec to add a new pod. My question is: Can we have set a dynamic rate limit on APIM? Something like: The current req/sec should not be > 50% of the previous requests?