Exam AZ-300 topic 16 question 11 discussion

C is correct. You can create an autoscale setting on a VM to use host-level metrics or guest OS-based metrics. ref : https://docs.microsoft.com/en-us/azure/virtual-machine-scale-sets/virtual-machine-scale-sets-autoscale-overview

I would go for C. There's nothing in here that states how we could benefit from using multiple profiles to better anticipate autoscaling which is why D seems too farfetched for me

D is the right option for workloads that have different usage rates during the year Best practices of Autoscale https://docs.microsoft.com/en-us/azure/azure-monitor/platform/autoscale-best-practices

C. Use only default diagnostics metrics to trigger autoscaling The use of the diagnostics metrics is the default when configuring autoscaling... (use of Diag metric to trigger scale). Diagnostics extension is not the same as diagnostics metric. The extension is an agent is used on VM for additional info. Answer D is something to confuse you! Note that the story was just for entertainment!

The question is tricky, it has taken me 5 minutes of reading to understand. And no, D is NOT CORRECT and here is why: The question tricks us into believing that there is a hardware performance problem when THERE IS NOT!!!! It says that the number of users vary throughout the year and that the infrastructure was resized to accommodate the peaks (probably thinking that we had a hardware performance issue) but the capacity WAS NOT UTILISED, which indicates that actually there was no problem with the hardware allocation. At this point, the info telling us that we do not have the source code becomes important. It may be a problem in the application. Having established the above, now look at the request. "You need to implement a Virtual machine scale set". A bit weird, isn't it? Just to implement a VMSS. Nothing else is requested. Well, in order to create a VMSS it is mandatory to define "a default diagnostic metric to trigger scaling" or answer C. Probably, it will never be triggered according to what we know. On the other hand, there is nothing on the statement pointing you towards multiple profiles.

I'd go for D. If you know you are busier at certain times if the year, makes sense to prepare for it, like create a profile "Christmas Sales" for the month December... Special Dates - I added a 'Product Launch Day' profile. I plan ahead for specific dates so my application is ready to handle the load due marketing announcements and when we put a new product in the application.* https://docs.microsoft.com/en-us/azure/azure-monitor/platform/autoscale-virtual-machine-scale-sets

I would go for D. Although using simple autoscale can be a solution for most of the scenarios, the fact that workload varies along the year makes the use of different autoscale profiles (each profile "tuned" for each situation) a more appropriate solution From https://docs.microsoft.com/en-us/azure/azure-monitor/platform/autoscale-overview You can have multiple profiles, which allow you to take care of different overlapping requirements. You can have different autoscale profiles for different times of day or days of the week, for example.

Should be D?