Edit

Share via

Troubleshoot disk performance using Microsoft Copilot in Azure

  • Article

Microsoft Copilot in Azure (preview) can help you troubleshoot issues with your disk performance when your application requires higher performance than what you have configured for your VM and disks. For more details about these issues, see Virtual machine and disk performance.

When you ask Microsoft Copilot in Azure questions about disk performance, it prompts you to select the VM and disks that are struggling with performance, along with the time period when the problems started. For best results, when specifying a time frame for the analysis, ensure that the disk and VM have been active during that entire period.

Using the information you provide, Copilot in Azure analyses your current configuration and performance metrics to determine whether your application is experiencing slowness due to reaching the configured performance limits for the VM or disk. It then provides a summary of the analysis and recommendations to resolve the performance issue, which you can apply directly in the Azure portal through Copilot in Azure's guided recommendations. The recommendations include a primary recommendation, which will be the least disruptive to your application, along with other possible options. For example, Copilot in Azure might determine that you can improve performance by upgrading your disks, enabling on-demand bursting, or adding another disk to your VM.

To diagnose performance issues when your application requires higher performance than what you have configured for your VM and disks, Copilot in Azure analyzes the following disk metrics:

  • Data Disk IOPS Consumed Percentage
  • Data Disk Bandwidth Consumed Percentage
  • OS Disk IOPS Consumed Percentage
  • OS Disk Bandwidth Consumed Percentage
  • VM Cached and Uncached IOPS Consumed Percentage
  • VM Cached and Uncached Bandwidth Percentage

Tip

The tasks and sample prompts listed here show a few of the areas where Microsoft Copilot in Azure (preview) can be especially helpful. However, this isn't a complete list of all the things you can do. We encourage you to experiment with your own prompts and see how Microsoft Copilot in Azure (preview) can help you manage your Azure resources and environment.

Important

Microsoft Copilot in Azure (preview) is currently in PREVIEW. See the Supplemental Terms of Use for Microsoft Azure Previews for legal terms that apply to Azure features that are in beta, preview, or otherwise not yet released into general availability.

Sample prompts

Here are a few examples of the kinds of prompts you can use to troubleshoot disk performance. Modify these prompts based on your real-life scenarios, or try additional prompts to meet your needs.

  • "Why is my disk slow?"
  • "Why is my VM attached to disks running slow?"
  • "Are there any bottlenecks affecting my disk performance?"
  • "What are the reasons for my disk's slow performance?"
  • "Help me with my VM-Disk performance."

Examples

When you ask Microsoft Copilot in Azure, "Why is my disk slow", Copilot runs an analysis of your disk and VM performance metrics to determine if your application performance is being capped due to requesting more IOPS or throughput than what is allotted for the virtual machines or attached disks. It starts by asking you to select the affected VM.

Screenshot of Microsoft Copilot in Azure responding to a prompt about slow disk performance.

After you select a VM, you're prompted to select one or more disks for Copilot in Azure to analyze.

Screenshot of Microsoft Copilot in Azure prompting to select disks to analyze.

Copilot in Azure also asks you when the issues began. You can enter an exact or approximate timeframe. For best results, be sure the VM and disk you selected have been active during the period you specify.

Screenshot of Microsoft Copilot in Azure asking to confirm the timeframe when disk performance issues began.

Copilot in Azure then shows you the VM and disks you selected and the metrics to be analyzed. After you confirm, Copilot in Azure runs the analysis to determine if your application performance is being capped due to requesting more IOPS or throughput than what is allotted for the virtual machines or attached disks.

Screenshot of Microsoft Copilot in Azure preparing to run an analysis of slow disk performance.

If Copilot in Azure detects a performance issue with your VM-Disk configuration due to hitting IOPS or throughput limits, it provides you with a summary of the analysis, a primary recommendation based on the least downtime to your application, and other recommendation options. You can also view metric details from the analysis metrics by selecting Show additional details. This option provides more information such as the VM IOPS/MBPS limits, total time period when limits were hit, and the top three time intervals when disk limits were hit. If your VM has caching enabled, the VM IOPS/MBPS limits shown here reflect your cached limits.

If you choose to enact any of the recommendation options provided by Copilot in Azure, you're directed to the location in the Azure portal where you can implement the recommendation. If you're dissatisfied with the recommendation, you can choose to submit a support request to get more assistance.

Next steps