Install KEDA on Kubernetes using Helm chart

[@hiandhra9971]

Hi Nilesh, your vedios are really helping me and if you could come up with more azure topics like, gitops, apim, etc, that would be really appreciated. Also after you install the keda with helm on your existing or new aks clusters. When I check keda triggers type. I mostly see azure app insights and cpu, metrics, azure monitor etc. but really which is recommended for aks cluster to use autoscaling?? We don’t use anything’s like rabbitmq, elastic etc etc.. we only use basic aks cluster but we want autoscaling enabled to save some cost. So do u recommend HPA or keda with cpu, memory triggers type ?

[@nilesh-gule]

Thank you for your feedback about my videos helping you in your learnings and also the suggestions for other topics. Currently there are 64 autoscaleres supported by KEDA as of this writing. keda.sh/docs/2.13/scalers/ You can use any of these based on your requirement. KEDA internally uses HPA for scaling up and down. The main advantage of using KEDA is that it can scale on external triggers which may not be part of your Kubernetes cluster like RabbitMQ or Kafka. I suggest you go through the list of scalers supported by KEDA and see which ones are applicable for your use case.

[@nlokesh1986]

Hi Sir, When we are upgrading KEDA to a latest version using Helm chart... do we need to remove existing Keda and delete the existing Scaled jobs?

[@nilesh-gule]

I have not tried updating the KEDA version. My understanding is that the upgrade process will take care of deleting the older version and replacing with the new one. Only issue will be when there are breaking changes between th eolder version and the new one.

[@nlokesh1986]

@@nilesh-gule I did the upgrade using Helm.. it went through without any issues..🙂