In this lab, you compare manual scaling of virtual machines to automatic scaling of virtual machines. You learn how to configure and resize a single virtual machine. You learn how create a virtual machine scale set and configure autoscaling.
This lab requires an Azure subscription. Your subscription type may affect the availability of features in this lab. You may change the region, but the steps are written using East US.
Your organization wants to explore deploying and configuring Azure virtual machines. First, you need to determine different compute and storage resiliency and scalability options you can implement when using Azure virtual machines. Next, you need to investigate compute and storage resiliency and scalability options that are available when using Azure virtual machine scale sets.
There are interactive lab simulations that you might find useful for this topic. The simulation lets you to click through a similar scenario at your own pace. There are differences between the interactive simulation and this lab, but many of the core concepts are the same. An Azure subscription is not required.
+ [Create a virtual machine in the portal](https://mslearn.cloudguides.com/en-us/guides/AZ-900%20Exam%20Guide%20-%20Azure%20Fundamentals%20Exercise%201). Create a virtual machine, connect and install the web server role.
+ [Deploy a virtual machine with a template](https://mslearn.cloudguides.com/en-us/guides/AZ-900%20Exam%20Guide%20-%20Azure%20Fundamentals%20Exercise%209). Explore the QuickStart gallery and locate a virtual machine template. Deploy the template and verify the deployment.
+ [Create a virtual machine with PowerShell](https://mslearn.cloudguides.com/en-us/guides/AZ-900%20Exam%20Guide%20-%20Azure%20Fundamentals%20Exercise%2010). Use Azure PowerShell to deploy a virtual machine. Review Azure Advisor recommendations.
+ [Create a virtual machine with the CLI](https://mslearn.cloudguides.com/en-us/guides/AZ-900%20Exam%20Guide%20-%20Azure%20Fundamentals%20Exercise%2011). Use the CLI to deploy a virtual machine. Review Azure Advisor recommendations.
## Task 1: Deploy zone-resilient Azure virtual machines by using the Azure portal
In this task, you will deploy two Azure virtual machines into different availability zones by using the Azure portal. Availability zones offer the highest level of uptime SLA for virtual machines at 99.99%. To achieve this SLA, you must deploy at least two virtual machines across different availabiliy zones.
1. Search for and select `Virtual machines` and, on the **Virtual machines** blade, click **+ Create**, and then select in the drop-down **+ Azure virtual machine**.
1. On the **Basics** tab of the **Create a virtual machine** blade, in the **Availability zone** drop down menu, place a checkmark next to **Zone 2**. This should select both **Zone 1** and **Zone 2**.
>**Note**: This will deploy two virtual machines in the selected region, one in each zone. You achieve the 99.99% uptime SLA because you have at least two VMs distributed across at least two zones. In the scenario where you might only need one VM, it is a best practice to still deploy the VM to a zone to ensure that the disk and corresponding resources are located in the same zone.
1. Click **Next: Disks >** and, on the **Disks** tab of the **Create a virtual machine** blade, specify the following settings (leave others with their default values):
| Setting | Value |
| --- | --- |
| OS disk type | **Premium SSD** |
| Enable Ultra Disk compatibility | **Unchecked** |
1. Click **Next: Networking >** take the defaults but do not provide a load balancer.
| Load balancing options | **None** |
1. Click **Next: Management >** and, on the **Management** tab of the **Create a virtual machine** blade, specify the following settings (leave others with their default values):
1. Click **Next: Monitoring >** and, on the **Monitoring** tab of the **Create a virtual machine** blade, specify the following settings (leave others with their default values):
| Setting | Value |
| --- | --- |
| Boot diagnostics | **Disable** |
1. Click **Next: Advanced >**, on the **Advanced** tab of the **Create a virtual machine** blade, review the available settings without modifying any of them, and click **Review + Create**.
1. On the **Review + Create** blade, click **Create**.
In this task, you will scale the compute for a virtual machine by adjusting its size to a different SKU. Azure provides flexibility in VM size selection so that you can adjust a VM for periods of time if it needs more (or less) compute and memory allocated. This concept is extended to disks, where you can modify the performance of the disk, or increase the allocated capacity.
In this task, you will deploy an Azure virtual machine scale set across availability zones. With individual VMs, you would need other automation to deploy and configure additional VMs if your application needs additional compute. VM Scale Sets reduce the administrative overhead of automation by enabling you to configure metrics or conditions that allow the scale set to automatically scale up or down the number of VMs in the set.
1. In the Azure portal, search for and select `Virtual machine scale sets` and, on the **Virtual machine scale sets** blade, click **+ Create**.
1. On the **Basics** tab of the **Create a virtual machine scale set** blade, specify the following settings (leave others with their default values) and click **Next : Spot >**:
| Already have a Windows Server license? | **Unchecked** |
>**Note**: For the list of Azure regions which support deployment of Windows virtual machines to availability zones, refer to [What are Availability Zones in Azure?](https://docs.microsoft.com/en-us/azure/availability-zones/az-overview)
1. On the **Networking** tab, click the **Create virtual network** link below the **Virtual network** textbox and create a new virtual network with the following settings (leave others with their default values). When finished, select **OK**.
1. On the **Edit network interface** blade, in the **NIC network security group** section, click **Advanced** and click **Create new** under the **Configure network security group** drop-down list.
1. On the **Create network security group** blade, specify the following settings (leave others with their default values):
| Setting | Value |
| --- | --- |
| Name | **vmss1-nsg** |
1. Click **Add an inbound rule** and add an inbound security rule with the following settings (leave others with their default values):
| Select a load balancer | **Create a load balancer** |
1. On the **Create a load balancer** page, specify the load balancer name and take the defaults. Click **Create** when you are done then **Next : Scaling >**.
| Setting | Value |
| --- | --- |
| Load balancer name | `vmss-lb` |
1. On the **Scaling** tab, specify the following settings (leave others with their default values) and click **Next : Management >**:
| Setting | Value |
| --- | --- |
| Initial instance count | `2` |
| Scaling policy | **Manual** |
1. On the **Management** tab, specify the following settings (leave others with their default values):
| Setting | Value |
| --- | --- |
| Boot diagnostics | **Disable** |
1. Click **Next : Health >**:
1. On the **Health** tab, review the default settings without making any changes and click **Next : Advanced >**.
1. On the **Advanced** tab, click **Review + create**.
1. On the **Review + create** tab, ensure that the validation passed and click **Create**.
1. Notice the **Scale mode** can be **Scale based on metrics** or **Scale to a specific instance count**. In scale sets with a small number of VM instances, increasing or decreasing the instance count may be best. In scale sets with a large number of VM instances, scaling based on metrics may be more appropriate.
1. Let's create a scale-out rule that automatically increases the number of VM instances. This rule scales out when the average CPU load is greater than 70% over a 10-minute period. When the rule triggers, the number of VM instances is increased by 20%. Click **Add** after making your selections.
1. Let's create a rule that decreases the number of VM instances in a scale set. The number of instances should decrease when the average CPU load drops below 30% over a 10-minute period. When the rule triggers, the number of VM instances is decreased by 20%. Adjust the settings, then select **Add**.
1. When your autoscale rules are applied, instance limits make sure that you do not scale out beyond the maximum number of instances, or scale in beyond the minimum number of instances.
1. Sign in to the Azure portal - `https://portal.azure.com`.
1. Use the menu to launch a **Cloud Shell** session. Alternately, navigate directly to `https://shell.azure.com`.
1. If necessary configure the Cloud Shell. Be sure to select **PowerShell**.
1. Run the following command to create a virtual machine. When prompted, provide a username and password for the VM. While you wait check out the [New-AzVM](https://learn.microsoft.com/powershell/module/az.compute/new-azvm?view=azps-11.1.0) command reference for all the parameters associated with creating a virtual machine.
```powershell
New-AzVm `
-ResourceGroupName 'az104-rg8' `
-Name 'myPSVM' `
-Location 'East US' `
-Image 'Win2019Datacenter' `
-Zone '1' `
-Size 'Standard_D2s_v3'
-Credential '(Get-Credential)' `
1. Once the command completes, use **Get-AzVM** to list the virtual machines in your resource group.
```powershell
Get-AzVM `
-ResourceGroupName 'az104-rg8'
-Status
1. Verify your new virtual machine is listed and the **Status** is **Running**.
>**Did you know?** When you use Azure to stop your virtual machine, the status is *deallocated*. This means that any non-static public IPs are released, and you stop paying for the VM’s compute costs.
## Task 6: Create a virtual machine using the CLI (option 2)
1. Run the following command to create a virtual machine. When prompted, provide a username and password for the VM. While you wait check out the [az vm create](https://learn.microsoft.com/cli/azure/vm?view=azure-cli-latest#az-vm-create) command reference for all the parameters associated with creating a virtual machine.
1. Use **az vm show** to ensure the **powerState** is **VM deallocated**.
>**Did you know?** When you use Azure to stop your virtual machine, the status is *deallocated*. This means that any non-static public IPs are released, and you stop paying for the VM’s compute costs.
If you are working with your own subscription take a minute to delete the lab resources. This will ensure resources are freed up and cost is minimized. The easiest way to delete the lab resources is to delete the lab resource group.
+ In the Azure portal, select the resource group, select **Delete the resource group**, **Enter resource group name**, and then click **Delete**.
+ Using Azure PowerShell, `Remove-AzResourceGroup -Name resourceGroupName`.
+ Using the CLI, `az group delete --name resourceGroupName`.
