Hey everyone, I’m trying to keep an eye on the performance of some VMs and VMSS instances on Azure, especially when it comes to: Average memory usage (RAM) Disk and network usage Average CPU usage Peak RAM and CPU usage (like the max spikes) Ideally, I’d love to see this info with timestamps, so I can track when those usage spikes happen. The reason I’m looking into this is that a few of my VMs have been throwing some errors lately, and I’m guessing it might be related to the sizing (RAM, CPU). Having these metrics would really help me figure out if that’s the case and help with any adjustments. Does anyone know the best way to gather this data for analysis? I’d appreciate any pointers! Thanks a lot!

Hi João Vitor do Prado Maia , Thank you for reaching out to the Microsoft Q&A platform. Based on your question, we understand that you want to know about the performance of some VMs and VMSS instances on Azure. VM insights monitors the performance and health of your virtual machines and virtual machine scale sets, including their running processes and dependencies on other resources. It can help deliver predictable performance and availability of vital applications by identifying performance bottlenecks and network issues and can also help you understand whether an issue is related to other dependencies. Please find the attached documents for your reference: Overview of VM insights Enable VM Insights overview Analyze metrics with Azure Monitor metrics explorer VM insights stores its data in Azure Monitor Logs, which allows it to deliver powerful aggregation and filtering and to analyze data trends over time. You can view this data in a single VM from the virtual machine directly, or you can use Azure Monitor to deliver an aggregated view of multiple VMs. If an answer has been helpful, please consider accepting the answer and "Upvote" to help increase visibility of this question for other members of the Microsoft Q&A community.

Monitoring VM and VMSS Performance on Azure

Accepted answer

Mahesh Goud Juvvadi 1,070 Reputation points Microsoft Vendor

2024-09-10T22:22:44.65+00:00

Hi João Vitor do Prado Maia,

Thank you for reaching out to the Microsoft Q&A platform.

Based on your question, we understand that you want to know about the performance of some VMs and VMSS instances on Azure.

VM insights monitors the performance and health of your virtual machines and virtual machine scale sets, including their running processes and dependencies on other resources. It can help deliver predictable performance and availability of vital applications by identifying performance bottlenecks and network issues and can also help you understand whether an issue is related to other dependencies.

Please find the attached documents for your reference:

Overview of VM insights

Enable VM Insights overview

Analyze metrics with Azure Monitor metrics explorer

VM insights stores its data in Azure Monitor Logs, which allows it to deliver powerful aggregation and filtering and to analyze data trends over time. You can view this data in a single VM from the virtual machine directly, or you can use Azure Monitor to deliver an aggregated view of multiple VMs.

If an answer has been helpful, please consider accepting the answer and "Upvote" to help increase visibility of this question for other members of the Microsoft Q&A community.
Please sign in to rate this answer.

1 person found this answer helpful.
James Bisiar 160 Reputation points

2024-09-10T21:12:25.8866667+00:00

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:

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.

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.

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:

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.

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.

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.

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:

Go to Azure Monitor > Workbooks.

Choose a template or create a custom workbook to track metrics like CPU, memory, disk, and network usage.

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:

Go to VMSS > Scaling in the Azure portal.

Define auto-scaling rules based on metrics such as CPU or memory usage.

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!

razec18 155 Reputation points

2024-09-11T04:37:04.4+00:00

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!

razec18 155 Reputation points

2024-09-11T05:35:53.8466667+00:00

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!

Mahesh Goud Juvvadi 1,070 Reputation points Microsoft Vendor

2024-09-11T19:27:33.46+00:00

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://learn.microsoft.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.

Mahesh Goud Juvvadi 1,070 Reputation points Microsoft Vendor

2024-09-12T19:00:39.02+00:00

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.

razec18 155 Reputation points

2024-09-12T20:34:13.89+00:00

Hey!

I did, I will try to do this today and tomorrow Ill give an update on this, ok?

razec18 155 Reputation points

2024-09-12T22:31:44.22+00:00

Mahesh Goud Juvvadi

I enabled there the performance diagnostics, but could not find exactly the resources / services that are using the most quantity of RAM / CPU

Is it possible?

Mahesh Goud Juvvadi 1,070 Reputation points Microsoft Vendor

2024-09-13T20:06:18.7366667+00:00

Hello João Vitor do Prado Maia,

To identify the services or processes using the most CPU, RAM, and other resources on your Azure VMs, you can follow the below steps:

Azure Monitor can help you track CPU, memory, and other performance metrics over time:

VM Insights: Enable this to get detailed insights, including CPU, memory, disk, and network usage.

Metrics Explorer: Allows you to create charts to monitor performance.

Log Analytics: You can query data to find specific resource usage trends. For example:

Perf | where (CounterName == "% Processor Time" or CounterName == "Working Set") and ObjectName == "Process" | summarize avg(CounterValue) by bin(TimeGenerated, 5m), Computer, InstanceName, CounterName | order by avg_CounterValue desc

Install the Azure Diagnostics Extension on your VMs to collect performance data, including CPU and memory usage, as well as process-specific information. Once enabled You can view process-level data from the Logs section in Azure Monitor.

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.

razec18 155 Reputation points

2024-09-17T03:19:19.4333333+00:00

Ill try this out and let you know.

I can do this with VM and VMSS right?

razec18 155 Reputation points

2024-09-17T03:19:54.4033333+00:00

Hi, I can do this both with VM and VMSS right?

Mahesh Goud Juvvadi 1,070 Reputation points Microsoft Vendor

2024-09-18T19:18:34.6233333+00:00

Hello João Vitor do Prado Maia,

Yes, you can check with both VM and VMSS.

If you are facing any issue do lets us know in the comments.
Sign in to comment

Use comments to ask for clarification, additional information, or improvements to the question.

Share via

Monitoring VM and VMSS Performance on Azure

1. Azure Monitor and Metrics

Steps:

Important Metrics to Monitor:

2. Log Analytics with Azure Monitor

Steps:

3. Azure Monitor Workbooks

Steps:

4. Alerts and Auto-Scaling (for VMSS)

Steps:

5. Third-Party Monitoring Tools

Conclusion

0 additional answers

Your answer