Editar

Compartilhar via


Monitor Azure Local with Azure Monitor Metrics

Applies to: Azure Local, version 23H2

This article describes how to monitor your Azure Local system with Azure Monitor Metrics. It also provides a comprehensive list of metrics collected for compute, storage, and network resources in Azure Local.

When you have critical applications and business processes that rely on Azure resources, it's important to monitor those resources for their availability, performance, and operation. The integration of Azure Monitor Metrics with Azure Local enables you to store numeric data from your clusters in a dedicated time-series database. This database is automatically created for each Azure subscription. Use metrics explorer to analyze data from your Azure Local system and assess its health and utilization.

Take a few moments to watch the video walkthrough on creating metric charts in metrics explorer and alerts using Metrics:

Benefits

Here are the benefits of using Metrics for Azure Local:

  • No additional cost. These metrics are standard, out-of-the-box features that are automatically collected and provided to you at no extra cost.

  • Near real-time insights. You have the capability to observe out-of-the-box metrics and correlate trends using near real-time data.  

  • Customization. You can create your own graphs and customize them through aggregation and filter functionality. The task of saving and sharing your metric charts via Excel, workbooks, or sending them to Grafana is straightforward.

  • Custom alert rules. You can write custom alert rules on the metrics to efficiently monitor the health of your Azure Local system.

Prerequisites

Here are the prerequisites of using Metrics for Azure Local:

  • You must have access to an Azure Local, version 23H2 (Build version: 2311) or later system that's deployed, registered, and connected to Azure.

  • The AzureEdgeTelemetryAndDiagnostics extension must be installed to collect telemetry and diagnostics information from your Azure Local system. For more information about the extension, see Azure Local telemetry and diagnostics extension overview.

Monitor Azure Local through the Monitoring tab

In the Azure portal, you can monitor platform metrics of your cluster by navigating to the Monitoring tab on your cluster's Overview page. This tab offers a quick way to view graphs for different platform metrics. You can select any of the graphs to further analyze the data in metrics explorer.

Follow these steps to monitor platform metrics of your system in the Azure portal:

  1. Go to your Azure Local cluster resource page and select your cluster.

  2. On the Overview page of your cluster, select the Monitoring tab.

    Screenshot showing the Monitoring tab for your cluster.

  3. On the Platform metrics pane, review the graphs displaying platform metrics. To know the metrics that Azure Monitor collects to populate these graphs, see Metrics for the Monitoring tab graphs.

    • At the top of the pane, select a duration to change the time range for the graphs.

    • Select the See all metrics link to analyze metrics using metrics explorer. See Analyze metrics.

    • Select any of the graphs to open them in metrics explorer to drill down further or to create an alert rule. See Create metrics alerts.

      Screenshot showing the platform metrics for your cluster.

Analyze metrics

You can use metrics explorer to interactively analyze the data in your metric database and chart the values of multiple metrics over time. To open the metrics explorer in the Azure portal, select Metrics under the Monitoring section.

Screenshot showing the Select a scope pane.

You can also access Metrics directly from the menu for the Azure Local services.

Screenshot of the Metrics page.

With Metrics, you can create charts from metric values and visually correlate trends. You can also create a metric alert rule or pin a chart to an Azure dashboard to view them with other visualizations. For a tutorial on using this tool, see Analyze metrics for an Azure resource.

Platform metrics are stored for 93 days, however, you can only query (in the Metrics tile) for a maximum of 30 days' worth of data on any single chart. To know more about data retention, see Metrics in Azure Monitor.

Analyze metrics for a specific cluster

Follow these steps to analyze metrics for a specific Azure Local cluster in the Azure portal:

  1. Go to your Azure Local cluster and navigate to the Monitoring section.

  2. To analyze metrics, select the Metrics option. Your cluster will already be populated in the scope section. Select the metric you want to analyze.

    Screenshot showing the metrics for your cluster.

    To create alerts, select the Alerts option and set up alerts as described in Create metric alerts.

What metrics are collected?

This section lists the platform metrics that are collected for the Azure Local cluster, the aggregation types, and the dimensions available for each metric. For more information about metric dimensions, see Multi-dimensional metrics.

Metrics for the Monitoring tab graphs

The following table lists the metrics that Azure Monitor collects to populate the graphs on the Monitoring tab:

Metrics Unit
Percentage CPU Percent
Network In/Sec BytesPerSecond
Network Out/Sec BytesPerSecond
Disk Read Bytes/Sec BytesPerSecond
Disk Write Bytes/Sec BytesPerSecond
Disk Read Operations/Sec CountPerSecond
Disk Write Operations/Sec CountPerSecond
Used Memory Bytes Bytes

Metrics for nodes

Metric Description Unit Default Aggregation Type Supported Aggregation Type Dimensions
Percentage CPU Percentage of processor time that isn't idle. Percent Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node
Percentage CPU Guest Percentage of processor time used for guest (virtual machine) demand. Percent Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node
Percentage CPU Host Percentage of processor time used for host demand. Percent Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node
Cluster node Memory Total The total physical memory of the node. Bytes Sum Minimum, Maximum, Average Cluster, Node
Cluster node Memory Available The available memory of the node. Bytes Maximum Minimum, Maximum, Average Cluster, Node
Cluster node Memory Used The used memory of the node. Bytes Maximum Minimum, Maximum Cluster, Node
Percentage Memory The allocated (not available) memory of the node. Percent Maximum Minimum, Maximum, Sum, Count Cluster, Node
Percentage Memory Guest The memory allocated to guest (virtual machine) demand. Percent Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node, LUN, VM
Percentage Memory Host The memory allocated to host demand. Percent Maximum Minimum, Maximum, Sum, Count Cluster, Node
Cluster node Csv cache Read Hit Cache hit PerSecond for read operations. CountPerSecond Maximum Minimum, Maximum, Sum, Count Cluster, Node, LUN
Cluster node Csv cache Read Hit rate Cache hit rate for read operations. Percent Maximum Minimum, Maximum, Sum, Count Cluster, Node, LUN
Cluster node Csv cache Read Miss Cache missPerSecond for read operations. CountPerSecond Maximum Minimum, Maximum, Sum, Count Cluster, Node, LUN
Cluster node Storage Degraded Total number of failed or missing drives in the storage pool. Bytes Sum Minimum, Maximum, Sum, Count Cluster, Node

Metrics for drives

Metric Description Unit Default Aggregation Type Supported Aggregation Type Dimensions
Physicaldisk Read Operations/Sec Number of read operations per second completed by the drive. CountPerSecond Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node, LUN
Physicaldisk Write Operations/Sec Number of write operations per second completed by the drive. CountPerSecond Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node, LUN
Physicaldisk Read and Write Operations/Sec Total number of read or write operations per second completed by the drive. CountPerSecond Sum Minimum, Maximum, Average, Sum, Count Cluster, Node, LUN
Physicaldisk Read Bytes/Sec Quantity of data read from the drive per second. BytesPerSecond Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node, LUN
Physicaldisk Write Bytes/Sec Quantity of data written to the drive per second. BytesPerSecond Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node, LUN
Physicaldisk Read and Write Total quantity of data read from or written to the drive per second. BytesPerSecond Sum Minimum, Maximum, Average, Sum, Count Cluster, Node, LUN
Physicaldisk Latency Read Average latency of read operations from the drive. Seconds Maximum Minimum, Maximum, Average, Sum Cluster, Node, LUN
Physicaldisk Latency Write Average latency of write operations to the drive. Seconds Maximum Minimum, Maximum, Average, Sum Cluster, Node, LUN
Physicaldisk Latency Average Average latency of all operations to or from the drive. Seconds Maximum Minimum, Maximum, Average, Sum Cluster, Node, LUN
Physicaldisk Capacity Size Total The total storage capacity of the drive. Bytes Sum Minimum, Maximum, Average Cluster, Node, LUN
Physicaldisk Capacity Size Used The used storage capacity of the drive. Bytes Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node, LUN

Metrics for network adapters

Metric Description Unit Default Aggregation Type Supported Aggregation Type Dimensions
Network In/Sec Rate of data received by the network adapter. Bytes Per Second Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node, Network Adapter, LUN
Network Out/Sec Rate of data sent by the network adapter. Bytes Per Second Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node, Network Adapter, LUN
Network Total/Sec Total rate of data received or sent by the network adapter. Bytes Per Second Sum Minimum, Maximum, Average, Sum, Count Cluster, Node, Network Adapter, LUN
Netadapter Bandwidth Rdma Inbound Rate of data received over RDMA by the network adapter. Bytes Per Second Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node, Network Adapter, LUN
Netadapter Bandwidth Rdma Outbound Rate of data sent over RDMA by the network adapter. Bytes Per Second Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node, Network Adapter, LUN
Netadapter Bandwidth Rdma Total Total rate of data received or sent over RDMA by the network adapter. Bytes Per Second Sum Minimum, Maximum, Sum, Count Cluster, Node, Network Adapter, LUN

Metrics for VHDs

Metric Description Unit Default Aggregation Type Supported Aggregation Type Dimensions
VHD Read Operations/Sec Number of read operations per second completed by the virtual hard disk. Count Per Second Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node, VHD
VHD Write Operations/Sec Number of write operations per second completed by the virtual hard disk. Count Per Second Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node, VHD
VHD Read and Write Operations/Sec Total number of read or write operations per second completed by the virtual hard disk. Count Per Second Sum Minimum, Maximum, Sum, Count Cluster, Node, VHD
VHD Read Bytes/Sec Quantity of data read from the virtual hard disk per second. Bytes Per Second Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node, VHD
VHD Write Bytes/Sec Quantity of data written to the virtual hard disk per second. Bytes Per Second Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node, VHD
VHD Read and Write Bytes/Sec Total quantity of data read from or written to the virtual hard disk per second. Bytes Per Second Sum Minimum, Maximum, Sum, Count Cluster, Node, VHD
VHD Latency Average Average latency of all operations to or from the virtual hard disk. Seconds Maximum Minimum, Maximum, Average, Sum Cluster, Node, VHD
VHD Size Current The current file size of the virtual hard disk, if dynamically expanding. If fixed, the series isn't collected. Bytes Maximum Minimum, Maximum, Average Cluster, Node, Instance
VHD Size Maximum The maximum size of the virtual hard disk, if dynamically expanding. Bytes Maximum Minimum, Maximum, Average Cluster, Node, VHD

Metrics for VMs

Metric Description Unit Default Aggregation Type Supported Aggregation Type Dimensions
VM Percentage CPU Percentage the virtual machine is using of its host node's processor(s). Percent Maximum Minimum, Maximum, Sum, Count Cluster, Node, VM
VM Memory Assigned The quantity of memory assigned to the virtual machine. Bytes Sum Minimum, Maximum Cluster, Node, LUN, VM
VM Memory Available The quantity of memory that remains available, of the amount assigned. Bytes Maximum Minimum, Maximum, Sum, Count Cluster, Node, VM, LUN
VM Memory Used VM Memory Used Bytes Maximum Minimum, Maximum Cluster, Node, VM, LUN
VM Memory Maximum If using dynamic memory, this is the maximum quantity of memory that might be assigned to the virtual machine. Bytes Maximum Minimum, Maximum, Average Cluster, Node, LUN, VM
VM Memory Minimum If using dynamic memory, this is the minimum quantity of memory that might be assigned to the virtual machine. Bytes Minimum Minimum, Maximum, Average Cluster, Node, LUN, VM
VM Memory Pressure The ratio of memory demanded by the virtual machine over memory allocated to the virtual machine. Bytes Maximum Minimum, Maximum, Average Cluster, Node, LUN, VM
VM Memory Startup The quantity of memory required for the virtual machine to start. Bytes Maximum Minimum, Maximum, Average Cluster, Node, LUN, VM
VM Memory Total Total memory. Bytes Maximum Minimum, Maximum, Average Cluster, Node, VM, LUN
VM network adapter Network In/Sec Rate of data received by the virtual machine across all its virtual network adapters. Bits Per Second Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node, Virtual Network Adapter
VM network adapter Network Out/Sec Rate of data sent by the virtual machine across all its virtual network adapters. Bits Per Second Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node, Virtual Network Adapter
VM network adapter Network In and Out/Sec Total rate of data received or sent by the virtual machine across all its virtual network adapters. Bits Per Second Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node, Virtual Network Adapter

Metrics for volumes

Metric Description Unit Default Aggregation Type Supported Aggregation Type Dimensions
Disk Read Operations/Sec Number of read operations per second completed by this volume. Count Per Second Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node, LUN
Disk Write Operations/Sec Number of write operations per second completed by this volume. Count Per Second Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node, LUN
Disk Read and Write Operations/Sec Total number of read or write operations per second completed by this volume. Count Per Second Sum Minimum, Maximum, Sum, Count Cluster, Node, LUN
Disk Read Bytes/Sec Quantity of data read from this volume per second. Bytes Per Second Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node, LUN
Disk Write Bytes/Sec Quantity of data written to this volume per second. Bytes Per Second Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node, LUN
Disk Read and Write Total quantity of data read from or written to this volume per second. Bytes Per Second Sum Minimum, Maximum, Sum, Count Cluster, Node, LUN
Volume Latency Read Average latency of read operations from this volume. Seconds Maximum Minimum, Maximum, Average, Sum Cluster, Node, LUN
Volume Latency Write Average latency of write operations to this volume. Seconds Maximum Minimum, Maximum, Average, Sum Cluster, Node, LUN
Volume Latency Average Average latency of all operations to or from this volume. Seconds Maximum Minimum, Maximum, Sum Cluster, Node, LUN
Volume Size Total The total storage capacity of the volume. Bytes Sum Minimum, Maximum, Average Cluster, Node, LUN
Volume Size Available The available storage capacity of the volume. Bytes Maximum Minimum, Maximum, Average, Sum, Count Cluster, Node, LUN

To see in-depth information about how these metrics are collected, see Performance history for Storage Spaces Direct.

Next steps