alawadi.cloudDocs

Autoscaling and alerts

Scale an app automatically between a minimum and maximum number of instances, and get told the moment a service is struggling, both in the portal and by email.

Two tools, one goal

Your app should survive a traffic spike without you watching a dashboard at 3 a.m. You should hear about trouble before your visitors do. Two features cover that:

  • Autoscaling adjusts the number of running instances of an app between a minimum and a maximum you choose, based on its average CPU and memory load. More traffic, more instances; quiet again, back down.
  • Alerts watch a metric you pick (CPU, memory, response latency, or restarts) and notify you in the portal (and by email, once your address is verified) when it crosses a threshold you set.

Nothing scales that you didn't configure, and nothing fires that you didn't ask for. Both are configured per service, from the service's page in the portal or via the API.

Use autoscaling when your load varies: a shop before holidays, an API with daytime peaks. Use alerts everywhere: they cost nothing, work for apps and managed databases, and are the difference between "the platform told me" and "a customer told me".

On this page