Microsoft Boosts Fabric and Azure Data Services to Simplify AI Development
Microsoft has announced significant updates across its data platforms at the annual Build conference, focusing on making AI application development more straightforward and efficient for developers. These updates enhance Microsoft Fabric, Azure Cosmos DB, SQL Server, Power BI, PostgreSQL, and integrations with Databricks and SAP services on Azure.
Azure Cosmos DB Integrated into Microsoft Fabric
Microsoft Fabric, which unifies data and analytics services, now supports semistructured data through Azure Cosmos DB, currently in preview. Cosmos DB is a NoSQL and vector database designed to handle documents, text, emails, and other unstructured or semistructured data. This integration allows data to be accessible via OneLake, Fabric’s unified data lake storage, enabling real-time analytics such as sentiment analysis in chat applications.
The Fabric Real-Time Intelligence feature now includes a digital twin builder in preview. This no-code/low-code tool helps developers create and manage virtual models of physical or logical entities, useful for scenarios like manufacturing, logistics, and customer profiling.
Azure Cosmos DB is also being expanded within the Azure AI Foundry platform to support AI agents capable of remembering and resuming previous user interactions. This thread storage feature, now generally available, ensures conversations maintain context, improving the natural flow of user experience.
Power BI introduces a new full-screen Copilot to enable conversational querying across multiple reports and semantic models. This Copilot integrates directly with Microsoft Teams and Microsoft 365. Additionally, Fabric data agents built with Copilot Studio can be embedded into Microsoft 365 and Teams to automate routine tasks and enhance interaction with enterprise data.
SQL Server 2025 Adds Vector Database Features
The upcoming 2025 edition of SQL Server, now in public preview, includes native vector search capabilities to support AI workloads. This update allows connections with local or cloud AI models and integrates with frameworks like LangChain for large language models and Semantic Kernel for AI agents. It also supports real-time data mirroring into Fabric and deployment through Azure Arc, helping keep SQL Server relevant for AI-driven applications.
PostgreSQL Enhancements on Azure
PostgreSQL on Azure receives major improvements aimed at helping developers extract deeper insights from relational data. A new Visual Studio Code extension integrated with GitHub Copilot provides AI-assisted query writing and schema design tools.
DiskANN, a high-performance vector indexing algorithm, is now generally available in Azure Database for PostgreSQL, offering faster performance compared to pgvector. Semantic operators powered by large language models are also in preview, enabling generative AI reasoning directly within PostgreSQL.
Stronger Databricks and SAP Collaboration on Azure
Microsoft plans to enhance integration with SAP’s Business Data Cloud and SAP Databricks on Azure later this summer. This will allow developers to build AI applications using SAP business data with Microsoft’s first-party Databricks service, combining scalability and security with Azure integration.
Azure Databricks Unity Catalog tables can now sync in real-time with OneLake, simplifying data access and management across cloud services. Azure AI Foundry agents gain the ability to execute Databricks jobs and leverage the Databricks AI/BI Genie conversational interface for data retrieval and contextual responses.
These updates reflect Microsoft's commitment to providing developers with unified, efficient tools to build intelligent applications more easily on its cloud and data platforms.
Your membership also unlocks:
