Microsoft Corp. today debuted a new and integrated data analytics platform called Microsoft Fabric that brings together all of the data and analytics tools an organization needs to build the foundation of artificial intelligence.

The platform was announced at Microsoft’s annual developer conference Build 2023 running today through Wednesday in Seattle. It integrates platforms such as Data Factory, Synapse and Power BI into a single, unified software-as-a-service product.

According to Microsoft, it will replace those disparate systems with a simpler, easier to manage and cost-effective integrated platform for companies looking to build and integrate AI into their technology stacks. It bundles all of the tools required by data professionals in one place, including data integration, data engineering, data warehousing, data science, real-time analytics, applied observability and business intelligence.

Microsoft said Fabric will make life much easier in a world that’s awash in data that’s generated by people’s devices, applications and interactions. Although organizations have already effectively harnessed much of this data to transform digitally and gain competitive advantages, there’s a need to simplify things as generative AI and large language models rise to the fore.

Services such as Azure OpenAI have enabled companies to create all manner of cutting edge AI experiences to make people more productive. But building such experiences is challenging as it requires a steady stream of clean data and a highly integrated analytics system. Most companies lack this, and instead have to contend with a labyrinth of disconnected tools and services, meaning AI development becomes both time consuming and extremely costly.

Microsoft Fabric is designed to change this, allowing organizations to use a single product that provides all of the capabilities their developers need to extract insights from data and make it available to AI or end users. At launch, Fabric supports seven core workloads, including Data Factory, which provides more than 150 connectors to popular cloud and on-premises data sources with drag-and-drop functionality.

It also supports Synapse data engineering, Synapse data science, Synapse data warehousing and Synapse real-time analytics, as well as visualization capabilities from Power BI. Finally, it supports real-time data detection and monitoring through Data Activator. All of these capabilities are available now in preview, Microsoft said.

To make life even easier, Microsoft Fabric will also integrate its own copilot tool, similar to something like GitHub Copilot. Available in preview soon, this will allow users to interact with Fabric using natural language commands and a chatlike interface, making it easier to generate code and queries, create AI plugins, enable custom Q&A, create visualizations and more.

Microsoft Fabric is built atop an open data lake platform called OneLake, which acts as a single source of truth and eliminates the need to extract, move or replicate data. Through OneLake, Microsoft said, Fabric also enables persistent data governance and a single capacity pricing model that scales as usage grows, while its open nature removes the risk of proprietary lock-in.

In addition to easing AI development tasks, Microsoft Fabric will help every user to harness the power of data, Microsoft said. The platform natively integrates with Microsoft 365 applications such as Microsoft Excel. As a result, someone using Excel can directly discover and analyze data from OneLake and generate a Power BI report in a single click.

Alternatively, someone using Microsoft Teams can use Fabric to bring data directly into their chats, channels, meetings and presentations, Microsoft said. Alternatively, a sales person using Dynamics 365 can use Fabric and OneLake to unlock insights on customer relationships, business processes and more.

Azure data updates

Microsoft Fabric was the headline among a slate of data-related updates announced at Build today. The company also announced a host of new capabilities in Power BI aimed at increasing user’s productivity.

The biggest one is Copilot for Power BI, available in preview now, which makes it easier to create reports or narrative summaries based on Power BI data in seconds. Users can also ask questions about their data in their natural language to generate answers, charts and visualizations.

Power BI Direct Lake, meanwhile, is a new storage mode that helps avoid data replication, while Power BI Desktop Developer Mode enables developer-centric workflows for Power BI datasets and reports through Git integration.

Microsoft’s cloud database service Azure Cosmos DB received a variety of updates that improve developer productivity and optimize costs. These include a new Burst Capacity option that’s said to improve performance for developers by making better use of the database’s idle throughput capacity to handle traffic spikes.

Microsoft claims that databases using standard provisioned throughput with burst capacity enabled will be able to maintain performance during short bursts when requests exceed the throughput limit. That, the company added, gives customers a “cushion” if they’re under-provisioned and reduces the number of rate-limited requests.

Other capabilities for Cosmos DB include hierarchical partition keys for more efficient partitioning strategies, materialized views for Cosmos DB for NoSQL, and .NET and Java SDK telemetry and app insights.

Finally, Cosmos DB is being updated with hyperscale pools, a shared resource model for Hyperscale databases that’s now in preview. Developers can build and manage new apps in the cloud and scale multiple databases that have varying and unpredictable usage demands.

Image: kjpargeter/Freepik