Reading the service limits we're unsure if azure functions my scale for our scenario. https://zcusa.951200.xyz/en-us/azure/azure-functions/functions-scale#service-limits
Simplified scenario:
We have a queue triggered function, let's call it SyncData(), that will do lots of syncrone api calls (https, same endpoint). The total duration of the api calls performed by SyncData() will vary but worst case it can take up to 4mins. (due to different api responsetimes)
Now we need to be sure how Azure Functions will scale in case we get a burst of "worst case" messages on the queue (that is messages that will cause SyncData() to take 4 min.
We're currently running the S3 plan.
Our understanding of the scaling:
- Theres no limit on outgoing connections, so this will not be a problem
- We can configure the manual scaling to 10 instances (functions hosts)
How can we figure out how many SyncData() invocations there will be running in parallell? (the actual work beeing done in SyncData() is not cpu intensive, mainly waiting for the current api call to complete)
How can we monitor the instance count and invocations count in parallell?
requests
| distinct cloud_RoleInstance
The about gives us total instance count, but how to get number of parallell invocations?
Thanks for any help