Skip to content

Latest commit

 

History

History
416 lines (324 loc) · 7.17 KB

azure-machine-learning-integration.mdx

File metadata and controls

416 lines (324 loc) · 7.17 KB
Raw
title tags metaDescription redirects freshnessValidatedDate
Azure Machine Learning monitoring integration
Integrations
Microsoft Azure integrations
Azure integrations list
New Relic's Microsoft Azure Machine Learning integration: what data it reports and how to enable it.
/docs/integrations/microsoft-azure-integrations/azure-integrations-list/azure-machine-learning-integration
/docs/azure-azure_machinelearning-integration
never

New Relic's integrations include an integration for reporting your Azure Machine Learning data to New Relic. This document explains how to activate this integration and describes the data that can be reported.

Activate integration [#activate]

To enable the integration follow standard procedures to activate your Azure service in New Relic.

Configuration and polling [#polling]

You can change the polling frequency and filter data using configuration options.

Default polling information for the Azure Machine Learning integration:

  • New Relic polling interval: 5 minutes

Find and use data [#find-data]

To find your integration data, go to one.newrelic.com > All capabilities > Infrastructure > Azure and select an integration.

You can query and explore your data using the following event type:

  <th>
    Event Type
  </th>

  <th>
    Provider
  </th>
</tr>

  <td>
    `AzureMachineLearningWorkspaceSample`
  </td>

  <td>
    `AzureMachineLearningWorkspace`
  </td>
</tr>
Entity
Workspace

For more on how to use your data, see Understand and use integration data.

Metric data [#metrics]

This integration collects Azure Machine Learning data for Workspace.

Machine Learning Workspace data

  <th style={{ width: "150px" }}>
    Unit
  </th>

  <th>
    Description
  </th>
</tr>

  <td>
    Count
  </td>

  <td>
    Number of runs completed successfully for this workspace
  </td>
</tr>

<tr>
  <td>
    `startedRuns`
  </td>

  <td>
    Count
  </td>

  <td>
    Number of runs started for this workspace
  </td>
</tr>

<tr>
  <td>
    `failedRuns`
  </td>

  <td>
    Count
  </td>

  <td>
    Number of runs failed for this workspace
  </td>
</tr>

<tr>
  <td>
    `modelRegisterSucceeded`
  </td>

  <td>
    Count
  </td>

  <td>
    Number of model registrations that succeeded in this workspace
  </td>
</tr>

<tr>
  <td>
    `modelRegisterFailed`
  </td>

  <td>
    Count
  </td>

  <td>
    Number of model registrations that failed in this workspace
  </td>
</tr>

<tr>
  <td>
    `modelDeployStarted`
  </td>

  <td>
    Count
  </td>

  <td>
    Number of model deployments started in this workspace
  </td>
</tr>

<tr>
  <td>
    `modelDeploySucceeded`
  </td>

  <td>
    Count
  </td>

  <td>
    Number of model deployments that succeeded in this workspace
  </td>
</tr>

<tr>
  <td>
    `modelDeployFailed`
  </td>

  <td>
    Count
  </td>

  <td>
    Number of model deployments that failed in this workspace
  </td>
</tr>

<tr>
  <td>
    `totalNodes`
  </td>

  <td>
    Count
  </td>

  <td>
    Number of total nodes. This total includes some of Active Nodes, Idle Nodes, Unusable Nodes, Preempted Nodes, Leaving Nodes.
  </td>
</tr>

<tr>
  <td>
    `activeNodes`
  </td>

  <td>
    Count
  </td>

  <td>
    Number of Active nodes. These are the nodes which are actively running a job.
  </td>
</tr>

<tr>
  <td>
    `idleNodes`
  </td>

  <td>
    Count
  </td>

  <td>
    Number of idle nodes. Idle nodes are the nodes which are not running any jobs but can accept new job if available.
  </td>
</tr>

<tr>
  <td>
    `unusableNodes`
  </td>

  <td>
    Count
  </td>

  <td>
    Number of unusable nodes. Unusable nodes are not functional due to some unresolvable issue. Azure will recycle these nodes.
  </td>
</tr>

<tr>
  <td>
    `preemptedNodes`
  </td>

  <td>
    Count
  </td>

  <td>
    Number of preempted nodes. These nodes are the low priority nodes which are taken away from the available node pool.
  </td>
</tr>

<tr>
  <td>
    `leavingNodes`
  </td>

  <td>
    Count
  </td>

  <td>
    Number of leaving nodes. Leaving nodes are the nodes which just finished processing a job and will go to Idle state.
  </td>
</tr>

<tr>
  <td>
    `totalCores`
  </td>

  <td>
    Count
  </td>

  <td>
    Number of total cores
  </td>
</tr>

<tr>
  <td>
    `activeCores`
  </td>

  <td>
    Count
  </td>

  <td>
    Number of active cores
  </td>
</tr>

<tr>
  <td>
    `idleCores`
  </td>

  <td>
    Count
  </td>

  <td>
    Number of idle cores
  </td>
</tr>

<tr>
  <td>
    `unusableCores`
  </td>

  <td>
    Count
  </td>

  <td>
    Number of unusable cores
  </td>
</tr>

<tr>
  <td>
    `preemptedCores`
  </td>

  <td>
    Count
  </td>

  <td>
    Number of preempted cores
  </td>
</tr>

<tr>
  <td>
    `leavingCores`
  </td>

  <td>
    Count
  </td>

  <td>
    Number of leaving cores
  </td>
</tr>

<tr>
  <td>
    `quotaUtilizationPercent`
  </td>

  <td>
    Percent
  </td>

  <td>
    Percent of quota utilized
  </td>
</tr>

<tr>
  <td>
    `cpuUtilizationPercent`
  </td>

  <td>
    Percent
  </td>

  <td>
    CPU utilization
  </td>
</tr>

<tr>
  <td>
    `gpuUtilizationPercent`
  </td>

  <td>
    Percent
  </td>

  <td>
    GPU utilization
  </td>
</tr>
Metric
`completedRuns`