An administrator is configuring autoscaling for an Azure Virtual Machine Scale Set (VMSS) named `vmss-app`. The workload for the application follows a predictable schedule:
* During business hours (Monday through Friday, 08:00 to 18:00), the VMSS must have a minimum of 5 instances and a maximum of 10 instances, scaling dynamically based on CPU utilization.
* Outside of business hours and during weekends, the VMSS must scale down to a minimum of 2 instances and a maximum of 4 instances to optimize costs.
Which of the following configuration methods should the administrator implement to meet these requirements with the least administrative effort?
- AConfigure a single autoscale profile with a metric-based rule that scales out based on CPU usage and set the scale-in rule's cooldown period to 14 hours.
- BCreate two separate Virtual Machine Scale Sets and use an Azure Load Balancer to route traffic based on a custom time-of-day rule.
- Configure two autoscale profiles: one default profile with capacity limits of 2 to 4, and one recurring profile configured for weekdays between 08:00 and 18:00 with capacity limits of 5 to 10.Answer
- DDeploy an Azure Logic App that uses a recurrence trigger to update the VMSS instance count using the Azure CLI at 08:00 and 18:00 daily.
Answer
Configure two autoscale profiles: one default profile with capacity limits of 2 to 4, and one recurring profile configured for weekdays between 08:00 and 18:00 with capacity limits of 5 to 10.
Azure Monitor Autoscale allows you to define multiple profiles. A default profile applies when no other profiles are active (handling weekends and off-hours). A recurring profile can be configured for specific schedules (weekdays during business hours) and will take precedence when its recurrence conditions are met, allowing different capacity limits and scale rules to apply automatically. This is the native, least administrative effort method.
Step-by-Step Solution
Key Concept
Azure Monitor Autoscale supports multiple scale profiles (default and recurring/scheduled) to dynamically adapt VMSS capacity limits and rules to predictable workload patterns.