Microsoft’s Azure Stream Analytics Platform as a Service (PaaS) supports mission-critical customer applications with real-time insights. The new features announced by Microsoft focus on key tenets including ease of use, developer productivity and enterprise readiness.
The rollout of preview features commences November 4th with worldwide availability coming throughout November.
Key additions to Stream Analytics
Two key features are online scaling and C# custom de-serializers and extensibility with C# custom code.
Online scaling enables users to increase or decrease SU capacity of a running job without having to stop it, contributing to Microsoft’s commitment to provide long-running mission-critical pipelines through Stream Analytics.
With C# custom de-serializers, developers can now leverage the power of Azure Stream Analytics to process data in Protobuf, XML, or any custom format. Custom de-serializers in C# can then be used to de-serialize events received by Azure Stream Analytics.
Furthermore, developers creating Stream Analytics modules in the cloud or on IoT Edge can now write or reuse custom C# functions and invoke them right in the query through User Defined Functions.
According to Microsoft, this enables scenarios such as complex math calculations, importing custom ML models using ML.NET, and programming custom data imputation logic.
Azure Stream Analytics now offers full support for Managed Identity based authentication with Power BI. However, customers can continue to use Azure Active Directory User-based authentication model.
More features coming soon
Microsoft is also rolling out a preview of Azure Stream Analytics supported on Azure Stack via IoT Edge runtime beginning January 2020. This will offer users the ability to analyse ingress data from Event Hubs or IoT Hub on Azure Stack, and egress the results to a blob storage or SQL database, according to Microsoft.
Other updates include debugging for Visual Studio and the ability to local test live data in Visual Studio Code.
Microsoft has also announced a private preview for Azure Machine Learning, including real-time scoring with custom Machine Learning models.
The models are managed by the Azure Machine Learning service and hosted by Azure Kubernetes Service (AKS) or Azure Container Instances (ACI). According to Microsoft the workflow doesn’t require any additional code to be written.