Monitoring VM and VMSS Performance on Azure

To effectively monitor the performance of your Azure Virtual Machines (VMs) and Virtual Machine Scale Sets (VMSS), Azure provides several tools and features that can help you track and analyze key metrics such as CPU, memory, disk, and network usage, including peak and average values over time. Here’s how you can gather and monitor these metrics with timestamps for detailed analysis:

1. Azure Monitor and Metrics

Azure Monitor is the go-to solution for tracking and analyzing performance metrics for VMs and VMSS instances.

Steps:

1. Enable Monitoring for VMs and VMSS:
   • Go to the Azure portal, navigate to your VM or VMSS, and make sure that monitoring is enabled under Insights.
   • Azure Monitor automatically collects metrics like CPU, memory, disk, and network usage. To capture more detailed metrics, ensure you have enabled the Azure Diagnostics extension on your VMs.
2. Track Key Metrics:
   • Go to Monitor in the Azure portal.
   • Under Metrics, select the appropriate resource (VM or VMSS) and choose the metrics you want to monitor. Some of the most important metrics include:
     • Average memory usage (RAM): While Azure Monitor does not natively track memory usage, you can enable Guest OS metrics by installing the Azure Diagnostics extension or using Log Analytics.
     • Disk and network usage: These metrics are available under Disk Read/Write Bytes and Network In/Out.
     • Average CPU usage: Select the Percentage CPU metric to monitor average CPU usage.
     • Peak CPU and RAM usage: You can set up alerts to track when usage exceeds certain thresholds or visualize peaks by adjusting the graph to show maximum values.
3. View Metrics with Timestamps:
   • Use the Azure Monitor Metrics explorer to visualize your data over time with timestamps. This tool allows you to set custom time ranges, and you can apply aggregation (such as average or maximum) to spot performance spikes.

Important Metrics to Monitor:

• Percentage CPU: For tracking CPU usage and spikes.
• Network In/Out: To measure network traffic in and out of the VM.
• Disk Read/Write Bytes: For tracking disk usage.
• Guest OS metrics (RAM): To track memory usage (requires diagnostics extension).

2. Log Analytics with Azure Monitor

For more advanced data collection, including tracking memory usage (RAM), you can use Azure Monitor with Log Analytics.

Steps:

1. Enable Log Analytics Workspace:
   • Create a Log Analytics Workspace from the Azure portal if you don’t already have one.
   • Link your VM and VMSS instances to the workspace under their Diagnostics Settings.
2. Collect Guest OS Metrics (RAM):
   • In the diagnostics settings, enable performance counters to collect memory data:
     • For memory, track counters like \Memory\Available MBytes and \Memory\% Committed Bytes In Use.
3. Run Queries for Peak and Average Usage:
   • Use Kusto Query Language (KQL) in Log Analytics to retrieve specific metrics like average and peak RAM or CPU usage with timestamps. For example:

      Perf
      | where ObjectName == "Memory" and CounterName == "% Committed Bytes In Use"
      | summarize AvgMemoryUsage = avg(CounterValue), MaxMemoryUsage = max(CounterValue) by bin(TimeGenerated, 1h)
      | order by TimeGenerated desc
   

   This query will give you both average and peak memory usage over time, with timestamps.
4. Set Alerts:
   • You can set up alerts based on your queries or metrics, for example, to get notified when CPU or RAM usage exceeds a specific threshold.

3. Azure Monitor Workbooks

Azure Workbooks provide customizable dashboards that allow you to visualize metrics and logs across your VMs and VMSS instances.

Steps:

1. Go to Azure Monitor > Workbooks.
2. Choose a template or create a custom workbook to track metrics like CPU, memory, disk, and network usage.
3. Configure graphs and tables to show time-series data with timestamps and analyze the performance of your resources over time.

4. Alerts and Auto-Scaling (for VMSS)

For VM Scale Sets, you can configure auto-scaling based on metrics like CPU or memory usage to handle spikes in demand. You can also set up alerts for specific performance thresholds to detect potential issues.

Steps:

1. Go to VMSS > Scaling in the Azure portal.
2. Define auto-scaling rules based on metrics such as CPU or memory usage.
3. Set up alerts in Azure Monitor to notify you when thresholds (e.g., CPU > 80%) are exceeded.

5. Third-Party Monitoring Tools

If you need more advanced monitoring capabilities, there are third-party tools available, such as Datadog, Prometheus, or New Relic, which integrate with Azure and provide detailed tracking of resource usage, along with customizable alerts.

Conclusion

To summarize:

• Use Azure Monitor to track core metrics like CPU, disk, and network usage, with visualizations available in the Metrics explorer.
• Enable Log Analytics to gather detailed metrics on memory (RAM) and track specific usage patterns over time using KQL queries.
• Utilize Azure Workbooks for custom dashboards and time-series analysis.
• Set up alerts for peak usage or spikes that may be causing errors.

This setup will allow you to monitor the performance of your VMs and VMSS instances, identify usage spikes, and make informed decisions regarding resource sizing and scaling.

Let me know if you need more detailed steps on any of these solutions!

Hey there!
Thanks for the reply @Mahesh Goud Juvvadi !

I'm currently using VM Insights to monitor my VMs and VMSS, and it's been great for keeping track of CPU, RAM, and other metrics. However, I was wondering if VM Insights can also help me identify which services or processes are using the most resources (CPU, RAM, etc.) on my VMs.

This would be super helpful for me to troubleshoot some performance issues and optimize my resources better.

Any ideas or guidance on this? I really appreciate the help!

Thanks in advance!

Hello James Bisiar is it possible, as well, to see wich processes are consuming most of my resources as well? This would be good for me.

Thanks a lot for your attention to this!

Hello João Vitor do Prado Maia,

Thank you for sharing your feedback.

VM Insights can help you identify which services or processes are using the most resources on your VMs.

Please check that VM Insights is enabled for your VMs.

search for Monitor >> Virtual Machines >>Insights >> Performance.

Here you can see various performance metrics such as CPU, memory, disk, and network usage.

Choose the specific processes, you can use the Performance Diagnostics. This tool provides detailed insights into the processes running on your VM and their resource consumption.

In the Performance tab, select Performance Diagnostics. This will open a new blade where you can see a list of processes and their resource usage.

Please find the attached document for your reference: https://video2.skills-academy.com/en-us/azure/azure-monitor/vm/vminsights-performance

Please don’t forget to Accept Answer and hit Yes for "was this answer helpful" wherever the information provided helps you. This can be beneficial to other community members for remediation for similar issues.

Thank you.

Hi João Vitor do Prado Maia,

Just checking in to see if you had a chance to see my response to your question. Please tell us if it was helpful and feel free to reach out to us if you have any queries.

If you feel that your quires have been resolved, please accept the answer by clicking the "Upvote" and "Accept Answer" on the post.

Thank you.