Azure App Service Automatic Scaling

Hi gauravseth,

I have dreamt about this feature for years and it was the main reason I was looking to move to AKS for better scaling control.

You want your web apps within the same app service plan to scale differently and independently of each other.

How are individual web apps scaled out differently and independently from others on the same plan? Can you pelase describe a bit more on how this works?

If the underlying plan needs to be scaled out for one web app, wouldn't all the web apps on that plan get scaled out with it?

Considering you have some interesting meachanism for the above, then can you comment on the below scenario:

• I have 10 apps on plan X and 10 apps on plan Y, both plans have a minimum instance count of 2, hence we are paying for 4 instances total.
• If one app on each plan needs to be scaled out due to their high request volume, can these two apps be scaled out onto a shared plan to cut costs, therefore 5 instances total?

It would be awesome if app service plans could just be in some sort of cluster group where web apps just get shuffled and spread out between them as needed and I deploy to the cluster and not a plan.

Will this in any way affect Regional VNet integration with Route All Traffic enabled on the web apps?

Also, how come this wasn't mentioned in the https://azure.microsoft.com/en-us/updates/ channel, this seems like a super amazing feature and should be better advertized.

And lastly, any ballpark ETA on GA?

Thanks,

Kris

Kris we scale individual web apps based on incoming HTTP requests and best-case basis.