How to find metrics for compute instances within Azure Machine Learning

Question

I'm searching metrics like CPU Percentage or Memory for compute instances in Azure Machine Learning. Can I get them somewhere like for a normal Azure VM?
I'm confused about the metrics directly within the Azure Machine Learning Workspace like CpuUtilization on the Metric Namespace: Machine Learning Service. They don't seem to have a relation to the compute instances because when I have a high usage on a compute instance it doesn't shows anything in this metric.
so 2 main questions:

1. Where can I find metrics for compute instances running within Azure ML
2. To what exactly are those metrics within Azure machine learning service referring to?

Answer 1

1. Go to Azure Portal > Azure Monitor > Metrics.
2. Select "Virtual Machine" and choose your compute instance to view CPU and memory metrics.
3. Go to Azure Monitor > Logs.
4. Query performance data using KQL in the Perf table.
   Machine Learning Service Metrics: These often refer to the Azure ML service infrastructure, not individual compute instances. Use Azure Monitor for instance-level metrics.