Use Case: Deploying an On-Demand Fleet with a Custom Image Featuring Microsoft Edge
In this scenario, we’ll walk through deploying an on-demand fleet in AWS AppStream 2.0 and building a custom image using the Image Builder.
For an introduction to AWS AppStream 2.0 and its features, please read “Part 1”.
Here, in Part 2, we use ‘AppStream-WinServer’ as our base image. Using the Image Assistant, we install Microsoft Edge as part of our custom image. Once the custom image is built, we attach it to the on-demand fleet and associate it with our AppStream Stack to make the application available to end users.
Step 1: Setting Up the Image Builder
We begin by launching the Image Builder using, in our case, ‘AppStream-WinServer2022-06-17-2024’ as the base image. To deploy it, we need to choose ‘name’, ‘instance type’, and ‘network configuration’. This environment provides a Windows Server platform, perfect for deploying and testing applications before rolling them out to a production environment.

Once the Image Builder is up and running, we access it via the ‘Connect’ button. Essentially, it is a machine running the OS you chose during the setup (in our case ‘AppStream-WinServer2022-06-17-2024’). After connecting, it will redirect you to a page where you select which user you would use and what its purpose is. Pick the ‘Administrator’ user for now.
After that you get a desktop environment displayed, just like a standard Windows OS desktop you would see on a physical or virtual machine. Click the Image Assistant to configure the environment and install Microsoft Edge.
The Image Assistant process involves six steps, with the first being ‘Add App‘—where you select the applications to include in your custom image. In this case, we want to add Microsoft Edge. To do so, click ‘Add App’, which will open the standard Windows file explorer. Navigate to the Desktop and select Microsoft Edge. Once selected, the App Launch Settings menu will appear. Simply click ‘Save’ to confirm.
This completes the first step of the Image Assistant. Continue through the remaining five steps to finalize the image configuration and optimize it for streaming in AWS AppStream 2.0.
In the final step, when launching the custom image creation process, the Image Assistant takes a snapshot of the configured machine (which, in this case, is based on AppStream-WinServer2022-06-17-2024). This snapshot includes both the base operating system and any applications added during setup, such as Microsoft Edge. The newly created custom image will then appear in the AppStream 2.0 console under the Image Registry.
Step 2: Attaching the Custom Image to the On-Demand Fleet
Now that the custom image is created, the next step is to deploy it to an on-demand fleet. On-demand fleets are perfect for balancing performance and cost, scaling up as needed based on user demand.
Configuration Steps:
- Name: Name your fleet (e.g., test-new-image-test)
- Instance Type: Select the appropriate instance type
- Capacity, Session, Stream, and Network Settings: Configure these based on your requirements
Next, create an AppStream Stack, which defines the user’s streaming environment. Configure settings like name, network, session, and permissions during Stack creation. Once complete, link your on-demand fleet to this AppStream Stack.
Step 3: Verifying the Application Deployment
Once the on-demand fleet is up and running with the custom image attached and linked to the Stack, the next step is to test the setup by logging into AWS AppStream. If everything is configured correctly, the Microsoft Edge app should be visible and fully functional within the Windows Server environment we created.
To provide access to the application, as described in Part 1 of this guide, you’ll need to assign a user to the Stack. This can be done from the AWS console by navigating to the User Pool, selecting the appropriate user, clicking ‘Actions’, and attaching the Stack. Once completed, the user will have access to the application within AppStream.
You will get an email notification from AWS where you will get a ‘Login page: Link’ and a temporary password which you can use to access the associated apps you have for your user, right after you change the temporary password. (if you haven’t already).
This step will confirm that the deployment has been successful and that our users can stream the custom environment with ‘Microsoft Edge’ available directly from the AppStream session.
Conclusion
In this guide, we demonstrated how to create a custom image in AWS AppStream 2.0 using the Image Builder and deploy it to an on-demand fleet. By using AppStream-WinServer2022-06-17-2024 as the base image and installing Microsoft Edge via the Image Assistant, we customized the streaming environment to meet user needs. This approach showcases the flexibility of AWS AppStream for delivering Windows-based applications in a scalable, on-demand environment.