Microsoft announced a set of new features in the fabric on Tuesday. This allows users to easily access data stored on Oracle and Google Cloud platforms, allowing them to better help develop AI and Analytics applications.
This update also adds support for model context protocols, graph databases, and geospatial data features.
The new feature currently being previewed has been revealed in Fabcon Vienna, a user conference for Microsoft fabric users in the Austrian capital.
Released in November 2023, Microsoft Fabric is an AI-powered data management and analytics platform. It aims to consolidate seven data workloads, including integration, management, and analytics.
According to William McKnight, president of McKnight Consulting Group, these are valuable additions for platform users, given the new features allow Microsoft Fabric users to access their data and advance AI initiatives.
“New features in Microsoft Fabric represent a significant update that enhances its capabilities as a comprehensive enterprise-grade data platform,” he said. “The addition positions fabric as a critical component for customers who embrace AI and take on data modernization initiatives.”
Together, these updates represent the width and depth resources that Microsoft can apply to this space. Individually, none of these functions solved the world's hunger, not the earth's shattering – but they touch on many important issues that companies face when manipulating data.
David MenningerAnalysts, ISG Software Research
David Menninger, an analyst at ISG Software Research, similarly said new features that are not significantly differentiated from those of competitors such as Oracle and Databricks would be beneficial.
“Together, these updates represent the breadth and depth of resources that Microsoft can apply to this space,” he said. “Individually, none of these features are earth shattering. They didn't solve the world's hunger, but they touch on many of the important issues facing companies when manipulating data.”
Previous Microsoft Fabric updates include the addition of database tools to promote AI development and AI tools within the platform.
New Features
According to Arun Ulagaratchagan, corporate vice president of Azure Data, the core goal of Microsoft Fabric is to enable businesses to consolidate their entire data estate.
To that end, Microsoft Fabric features Onelake, a data lake where customers can store not only Microsoft data, but also data stored on platforms such as Amazon Redshift and Databricks. Onelake includes a feature called Mirroring that automatically generates replicas of data stored on another platform.
Mirroring allows users of multiple cloud data storage platforms to combine data, allowing them to be used together to notify AI and Analytics applications. The latest Microsoft Fabric Update adds mirroring of data stored in Oracle and Google Bigquery.
“The heart of Fabric Vision… unifies global data,” Uragarachagan said. “We use either shortcuts, connectors, or mirroring to neatly sweep away all our key data sources.”
Meanwhile, according to Menninger, eliminating extracts, conversions and loads (ETL) pipelines is a key trend in data management. Mirroring allows you to use your data without creating complex, costly pipelines. Therefore, Menninger pointed out that it is important to add Oracle and BigQuery mirroring.
“Zero Copy and Zero ETL are hot topics right now,” he said. “By extending these capabilities to include Oracle and BigQuery, we eliminate all efforts to maintain a pipeline that allows large amounts of data to be accessed much easier in a Microsoft environment, consolidating the data and performing that integration.”
Similar to the mirroring capabilities added to Onelake, Ulagaratchagan says fabric graphs and fabric maps aim to better integrate data.
Fabric Graphs are graph databases built on LinkedIn's graph technology, owned by Microsoft. Use neural networks to discover relationships between data points that go beyond the capabilities of traditional relational databases.
You can enable graph databases to allow data points to connect to multiple other things at the same time, but the relational database allows you to connect only one data point at a time. Graph databases are often used for fraud detection, recommendation engines, and risk assessment because they can discover impossible relationships in relational databases.
Fabric maps allow users to add geospatial contexts to AI agents, generative AI assistants, and analytic applications by combining streaming location data with mapping and modeling capabilities.
“The graph database was probably the last missing part of the fabric that has become a complete data platform,” Ulagaratchagan said.
McKnight similarly noted that, combined with additional mirroring capabilities, the new database has made Microsoft Fabric a more complete platform and a more complete platform for providing information to AI tools.
“These features enhance the ability to integrate and enrich fabric data foundations. [to] “We support the development of highly intelligent AI agents that understand business operations,” he said.
Fabric MCP is also intended to support agent AI development.
MCP is an open framework developed by AI vendor humanity in November 2024. Simplify development agents by standardizing how agents interact with data sources that they train, such as external sources such as company-specific data and large language models.
Many data management vendors are adding support for MCP, as agents are the current pioneer in AI development. Microsoft already offers MCP support on platforms such as Copilot Studio and Azure AI Foundry. Now, it's also done with cloth.
“Because Fabric is Microsoft's AI data platform, MCP support simply extends AI capabilities,” Ulagaratchagan said.
However, despite Microsoft's focus on fabric AI development, the capabilities of platforms offered by competitors such as Databricks and Snowflake have not yet matured, despite Microsoft's focus on fabric AI development.
“While Fabric has invested heavily in AI, it may not directly provide the same level of refinement or integration as its competitors to its analytical workflow,” he said. “DataBricks builds AI on an open lakehouse architecture and integrates AI directly into the analytics workflow. SnowFlake's… AI features are mature to provide 'easy buttons' for basic AI use cases. ”
Next Steps
Mirroring features, graphs, geospatial databases, and MCP support are all in preview, effectively configuring the fabric roadmap.
Micknight has proposed that beyond what Microsoft is planning, Tech Giant will do more to create a fabric AI development network that will extend beyond other Microsoft features.
“Microsoft has deepened its commitment to opening standards to nurture a wider developer ecosystem, developing more differentiated, sophisticated AI/ML capabilities and workloads, providing more detailed control and performance tuning options for power users,” he said. “In addition, we may use a better position for ease of use for core data management and AI preparation.
Menninger says that data fabric is an architecture that unifies enterprise data assets while Microsoft Fabric makes the concept of data fabric an enterprise-grade reality.
“The semantic model will be one of the next big frontiers,” he said. “They provide context for AI and BI. But it's not easy to create. … I hope we'll see more over time from Microsoft and others.”
Eric Avidon is a senior news writer at Informa TechTarget and a journalist with over 25 years of experience. He covers analytics and data management.
