How to scale azure function instances on premise?

[Insadem]

Hello, I deployed azure function app on vm, it works great.
I use azure storage as data storage provider and it works great too.

However, when I add another vm there's concurrency which leads to problems (such as processing data 2 times).
As far as I understand azure cloud scales azure app across multiple instances, but why I can't do the same on premises?

It's 100% not related to my code implementation, because I already tried to run it on azure cloud and it worked great.
My host.json:
{ "version": "2.0", "logging": { "fileLoggingMode": "debugOnly", "applicationInsights": { "samplingSettings": { "isEnabled": false, "initialSamplingPercentage": 0.0, "minSamplingPercentage": 0.0, "maxSamplingPercentage": 0.0 }, "enableLiveMetrics": false, "enableDependencyTracking": false, "enablePerformanceCountersCollection": false, "httpAutoCollectionOptions": { "enableHttpTriggerExtendedInfoCollection": false, "enableW3CDistributedTracing": false, "enableResponseHeaderInjection": false } } }, "concurrency": { "dynamicConcurrencyEnabled": true, "snapshotPersistenceEnabled": true }, "extensions": { "durableTask": { "hubName": "DurableFunctionsHub", "storageProvider": { "connectionStringName": "AzureWebJobsStorage", "trackingStoreNamePrefix": "DurableTask", "controlQueueBatchSize": 32, "controlQueueBufferThreshold": 256, "maxQueuePollingInterval": "00:00:01" }, "tracing": { "traceInputsAndOutputs": false, "traceReplayEvents": false }, "maxConcurrentActivityFunctions": 30, "maxConcurrentOrchestratorFunctions": 50, "maxEntityOperationBatchSize": 1000 } } }

[vigouredelaruse]

in case i have to answer this queston for points on the package i looked into it

• reviewed https://learn.microsoft.com/en-us/azure/azure-functions/functions-concurrency and can see that the concurrency settings matter in host.json
• reviewed scale controlle activity https://learn.microsoft.com/en-us/azure/azure-functions/event-driven-scaling?tabs=azure-cli
• reviewed keda documentation and see the claim is seamless autoscaling https://learn.microsoft.com/en-us/azure/azure-functions/functions-kubernetes-keda

note that the only on-prem configurations discussed here (https://learn.microsoft.com/en-us/azure/architecture/hybrid/azure-functions-hybrid) refer to docker/kubernetes/keda

having said that, i have yet to encounter concurrency issues on prem in a kubernetes/keda configuration, with rabbitmq triggers though