Introduction
Azure AI (Azure AI Foundry / Azure Machine Learning) is a managed platform for building, deploying, and operating machine learning models and generative AI applications. OpsRamp discovers and monitors Azure AI resources, including model deployments, online endpoints, online endpoint deployments, and workspace connections. It also collects Azure Monitor metrics from the Microsoft.MachineLearningServices namespace, providing visibility into resource health, performance, and usage.
Use the Azure Public Cloud integration to discover Azure AI resources and collect associated performance and operational metrics.
Setup
Before discovering Azure AI resources, ensure that the OpsRamp Microsoft Azure Integration is configured.
To enable Azure AI resource discovery, select Azure AI as a resource type in the Azure Integration Discovery Profile.
OpsRamp collects Azure Monitor metrics for the following resource types:
Microsoft.MachineLearningServices/workspaces/onlineEndpointsMicrosoft.MachineLearningServices/workspaces/onlineEndpoints/deployments
Ensure that the integration application is assigned the Monitoring Reader role on the target Azure subscription(s).
Supported resource types
| Resource Type (Native Type) | Resource Tag | Description |
|---|---|---|
| Azure AI Model Deployment | AZ_AI_DEPLOYMENT | Model deployments within an Azure AI and ML workspace. |
| Azure AI Online Endpoint | AZ_AI_ONLINE_ENDPOINT | Online (real-time) inference endpoints. |
| Azure AI Online Endpoint Deployment | AZ_AI_ONLINE_DEPLOYMENT | Deployments serving models behind an online endpoint. |
| Azure AI Workspace Connection | AZ_AI_CONNECTION | Connections from a workspace to external resources and services. |
Supported metrics
Online Endpoint
| OpsRamp Metric | Azure Metric | Metric Display Name | Unit | Aggregation Type | Description |
|---|---|---|---|---|---|
| azure_machinelearningservices_workspaces_onlineendpoints_ConnectionsActive | ConnectionsActive | Connections Active | count | Average | Number of active connections. |
| azure_machinelearningservices_workspaces_onlineendpoints_DataCollectionErrorsPerMinute | DataCollectionErrorsPerMinute | Data Collection Errors Per Minute | count | Average | Number of data collection errors per minute. |
| azure_machinelearningservices_workspaces_onlineendpoints_DataCollectionEventsPerMinute | DataCollectionEventsPerMinute | Data Collection Events Per Minute | count | Average | Number of data collection events per minute. |
| azure_machinelearningservices_workspaces_onlineendpoints_NetworkBytes | NetworkBytes | Network Bytes | Bps | Average | Network throughput in bytes per second. |
| azure_machinelearningservices_workspaces_onlineendpoints_NewConnectionsPerSecond | NewConnectionsPerSecond | New Connections Per Second | counts/s | Average | New connections established per second. |
| azure_machinelearningservices_workspaces_onlineendpoints_RequestLatency | RequestLatency | Request Latency | ms | Average | Request latency. |
| azure_machinelearningservices_workspaces_onlineendpoints_RequestLatency_P50 | RequestLatency_P50 | Request Latency P50 | ms | Average | 50th percentile request latency. |
| azure_machinelearningservices_workspaces_onlineendpoints_RequestLatency_P90 | RequestLatency_P90 | Request Latency P90 | ms | Average | 90th percentile request latency. |
| azure_machinelearningservices_workspaces_onlineendpoints_RequestLatency_P95 | RequestLatency_P95 | Request Latency P95 | ms | Average | 95th percentile request latency. |
| azure_machinelearningservices_workspaces_onlineendpoints_RequestLatency_P99 | RequestLatency_P99 | Request Latency P99 | ms | Average | 99th percentile request latency. |
| azure_machinelearningservices_workspaces_onlineendpoints_RequestsPerMinute | RequestsPerMinute | Requests Per Minute | count | Average | Number of requests per minute. |
Online Endpoint Deployment
| OpsRamp Metric | Azure Metric | Metric Display Name | Unit | Aggregation Type | Description |
|---|---|---|---|---|---|
| azure_machinelearningservices_workspaces_onlineendpoints_deployments_CpuMemoryUtilizationPercentage | CpuMemoryUtilizationPercentage | CPU Memory Utilization Percentage | % | Average | Percentage of memory used on an instance. |
| azure_machinelearningservices_workspaces_onlineendpoints_deployments_CpuUtilizationPercentage | CpuUtilizationPercentage | CPU Utilization Percentage | % | Average | Percentage of CPU utilization on an instance. |
| azure_machinelearningservices_workspaces_onlineendpoints_deployments_DataCollectionErrorsPerMinute | DataCollectionErrorsPerMinute | Data Collection Errors Per Minute | count | Average | The number of data collection events dropped per minute. |
| azure_machinelearningservices_workspaces_onlineendpoints_deployments_DataCollectionEventsPerMinute | DataCollectionEventsPerMinute | Data Collection Events Per Minute | count | Average | The number of data collection events processed per minute. |
| azure_machinelearningservices_workspaces_onlineendpoints_deployments_DeploymentCapacity | DeploymentCapacity | Deployment Capacity | count | Average | The number of instances in the deployment. |
| azure_machinelearningservices_workspaces_onlineendpoints_deployments_DiskUtilization | DiskUtilization | Disk Utilization | % | Average | Percentage of disk utilization on an instance. |
| azure_machinelearningservices_workspaces_onlineendpoints_deployments_GpuEnergyJoules | GpuEnergyJoules | GPU Energy in Joules | count | Average | Energy consumed on a GPU node during the interval, in joules. |
| azure_machinelearningservices_workspaces_onlineendpoints_deployments_GpuMemoryUtilizationPercentage | GpuMemoryUtilizationPercentage | GPU Memory Utilization Percentage | % | Average | Percentage of GPU memory utilization on an instance. |
| azure_machinelearningservices_workspaces_onlineendpoints_deployments_GpuUtilizationPercentage | GpuUtilizationPercentage | GPU Utilization Percentage | % | Average | Percentage of GPU utilization on an instance. |
| azure_machinelearningservices_workspaces_onlineendpoints_deployments_RequestLatency_P50 | RequestLatency_P50 | Request Latency P50 | ms | Average | The average P50 request latency. |
| azure_machinelearningservices_workspaces_onlineendpoints_deployments_RequestLatency_P90 | RequestLatency_P90 | Request Latency P90 | ms | Average | The average P90 request latency. |
| azure_machinelearningservices_workspaces_onlineendpoints_deployments_RequestLatency_P95 | RequestLatency_P95 | Request Latency P95 | ms | Average | The average P95 request latency. |
| azure_machinelearningservices_workspaces_onlineendpoints_deployments_RequestLatency_P99 | RequestLatency_P99 | Request Latency P99 | ms | Average | The average P99 request latency. |
| azure_machinelearningservices_workspaces_onlineendpoints_deployments_RequestsPerMinute | RequestsPerMinute | Requests Per Minute | count | Average | Number of requests sent to the online deployment within one minute. |