AVD Auto Scaling

AVD Autoscaling is a built in feature in AVD service to provide intelligent power management solution based on the user load and time(peak hours and non-peak hours). The AVD auto scaling was previously configured using Azure automation and Logic Apps. The autoscaling feature is now embedded in the Azure Portal under AVD service.

To configure Auto scaling in AVD service, navigate to AVD service in the Azure portal and click on “scaling plans”.

Click on “+Create” to initiate the scaling plan creation wizard.

Figure1:- Create Scaling Plan

In the basics tab, provide the mandatory information as described in the below table and Click Next

Parameter	Value
Subscription	<provide your subscription>
Resource Group	<RG where the scaling policy to be kept>
Name	Name of the Scaling plan
Location	Location where the metadata to be saved
Time zone	time zone where the scaling plan to be created. This is important as the Peak hours and non-peak hours will be calculated based on the time zone
Exclusion Tag	The exclusion tag which can be applied to exclude the session hosts from the scaling plan. this can be applied to session hosts which are in planned maintenance.

Figure 2:- Basics Option

In the Schedules tab, Click on “+Add schedule” to add the scaling schedule.

Figure 3:- Add schedule

A new wizard will be displayed where you can configure the scaling schedule.

In the General tab, Provide a name for your scaling schedule and select the days on which the scaling need to be run.

Figure 4:- General Schedule tab

Figure 5:- Repeat On Option in General tab

In the “Ramp up” tab, provide the information as described in the below table

Parameter	Value
Start time	<provide the ramp up time where the machines will get prepared for peak hours>
Load balancing algorithm	Select from either Breadth first/depth first algorithm
Minimum % of hosts	Mi. % of session hosts to be ramped up for the peak hours
capacity threshold	This is to provide the used host pool capacity that will be considered to evaluate whether to turn on/off virtual machines during the ramp-up and peak hours.

Figure 6:- Ramp Up Schedule configuration

In the “Peak Hours” tab, provide the start time of the business and the load balancing algorithm to be used during Peak hours.

Figure 7:- Peak Hours configuration

Click Next once the settings were configured.

In the “Ramp down” tab, provide the information as described in the table below. Click next once done.

Parameter	Value
Start time	Ramp down start time. Basically the time to prepare the machines for non-peak hours
Load balancing algorithm	Select from either Breadth first/depth first algorithm
Min. % of hosts	% of hosts to be kept untouched during the ramp down period
capacity threshold	percentage of used host pool capacity that will be considered to evaluate whether to turn on/off virtual machines during the ramp-down and off-peak hours.
Force logoff users	Select whether you can force logoff users or not
Delay time before user logoff	time in minutes to be wait before logging off the users from session host
Notification message	Message to be displayed to users before logging off from the session hosts

Figure 8:- Ramp down plan configuration

In the “Off-peak” hours tab, select the off peak start time and the load balancing algorithm to be used during the offpeak hours.

Figure 9:- Off peak hours configuration

In the “host pool assignments” tab, select the host pool to be applied with auto scaling policy and enable the autoscale option.

Figure 10:- Host pool assignments

Click Next and select the required tags. Click Next and Finish to complete the scheduling and scaling plan assignment.

Figure 11:- Scaling Option Summary