MLOps

Machine Learning and Analytics on Azure

Data Factory feeds a staging area; Azure Machine Learning trains, registers and serves models; Power BI and apps use the results.

From the AI and machine-learning service in NGT's Azure offering: enterprise and third-party data flow through Data Factory into a staging area, Azure Machine Learning trains and registers models, batch scores feed analytics and Power BI, and managed online endpoints answer applications in real time.

ingesttraining setsregisterdeploybatch scorespredictionsModel dataMicrosoft AzureData workloadStaging areaArtificial intelligenceAnalytical workloadFront endEnterprise dataThird-party metadataAzure Data FactoryData Lake StorageAzure SynapseAnalyticsAzure SQL DatabaseAzure MachineLearningtraining pipelinesModel registryManaged onlineendpointreal-time inferenceAzure SynapseAnalyticsscored dataPower BIWeb application

Point at a component to light its links; select it for what it does here and the practice that makes it right on Azure.

When to use it

  • Predictions such as anomaly or fraud detection on Azure data
  • Models that must be versioned, approved and monitored
  • Teams that already report in Power BI

How NGT delivers it

Built from code, reviewed at a gate by an NGT architect, and handed over with runbooks your team can run.

MLOps