Azure VM Scale Set Features Enhance Management, Flexibility

• Azure VM scale sets enable a set of VMs to scale out or in based on demand for the application they are delivering.
• New capabilities provide more control over scaling and make it easier to customize the VMs in a scale set.
• A new feature in preview can identify unhealthy VM instances and replace them to improve scale set reliability.

New capabilities recently added to Azure VM scale sets provide additional control over how sets are scaled in, offer more flexibility in the VM instances that can be added to a scale set, and can address faulty VM instances in a scale set.

Service Overview

Scale sets offer a deployment and management framework to deliver scalable applications running Linux or Windows Server on a set of identical Azure VMs. When a scale set is deployed, a set of VMs is initially created and traffic is distributed via a load balancer. Autoscale rules can be configured based on metrics from the VM instances in the set, such as CPU utilization, memory use, or data traffic. When application demand changes, the number of VM instances in the scale set can be increased (scaled out) to accommodate increased demand, or the number of VM instances can be decreased (scaled in) to accommodate lower demands, depending on the configured autoscale rules.