Migrating an On-Premises Third-Party Website to Azure - Best Practices

Implementing a Virtual Machine Scale Set Instance

Question

A company is migrating an existing on-premises third-party website to Azure. The website is stateless.

The company does not have access to the source code for the website. They have the original installer.

The number of visitors at the website varies throughout the year. The on-premises infrastructure was resized to accommodate peaks but the extra capacity was not used.

You need to implement a virtual machine scale set instance.

What should you do -

Answers

Explanations

Click on the arrows to vote for the correct answer

A. B. C. D.

C

In-guest VM metrics with the Azure diagnostics extension

The Azure diagnostics extension is an agent that runs inside a VM instance. The agent monitors and saves performance metrics to Azure storage. These performance metrics contain more detailed information about the status of the VM, such as AverageReadTime for disks or PercentIdleTime for CPU. You can create autoscale rules based on a more detailed awareness of the VM performance, not just the percentage of CPU usage or memory consumption.

https://docs.microsoft.com/en-us/azure/virtual-machine-scale-sets/virtual-machine-scale-sets-autoscale-overview

To efficiently manage the website that is stateless and varying visitors throughout the year, virtual machine scale set instances can be used. The scale set allows you to manage and automate the scaling of VMs on the Azure platform based on the defined rules.

The autoscaling option can be used to automatically adjust the number of VM instances as per demand. Autoscaling can be configured using various metrics, including CPU usage, memory usage, or network traffic. In this scenario, as the company has the original installer but not the source code, using an autoscale setting is the most suitable option for the company.

The correct answer is D. Use an autoscale setting to define more profiles that have one or more autoscale rules.

Option A, "Use a webhook to log autoscale failures," is incorrect because the webhook does not relate to scaling or managing the virtual machines.

Option B, "Use an autoscale setting to scale instances vertically," is incorrect because scaling vertically means resizing the VMs by adding more resources like CPU, memory, or disk. This may not be the most cost-effective or practical solution to handle varying website traffic.

Option C, "Use only default diagnostics metrics to trigger autoscaling," is incorrect because the default metrics may not be sufficient to trigger autoscaling based on the website traffic. It is recommended to define more custom metrics to scale the VMs more accurately.

Therefore, option D is the correct answer as it allows defining more profiles with one or more autoscale rules that can help manage and scale the VMs based on traffic demand. By defining more profiles, the company can have more precise control over the scale set, resulting in optimal resource utilization and cost savings.