Microsoft Fabric Stitches Together Data and Analytics

I have written before about the rising popularity of the data fabric approach for managing and governing data spread across distributed environments comprised of multiple data centers, systems and applications. I assert that by 2025, more than 6 in 10 organizations will adopt data fabric technologies to facilitate the management and processing of data across multiple data platforms and cloud environments. The data fabric approach is also proving attractive to vendors, including Microsoft, as a means of combining a variety of tools and platforms into consolidated platform offerings designed to provide a strategic approach to data management and data governance.

Microsoft is well-established as a provider of data platforms and analytics products, with a diverse portfolio that addresses databases, data integration, data management, data governance, business intelligence (BI), artificial intelligence (AI), and machine learning (ML) — both on premises and via its Azure cloud service. At its Build event for developers and engineers in May, the company announced the launch of a new offering that combines key elements of its data and analytics portfolio into a single product designed to facilitate and accelerate data-driven decision-making. Microsoft Fabric is a unified data and analytics platform designed to provide a single environment and experience for data integration, processing and analytics. Microsoft Fabric combines capabilities of existing products, including Azure Data Factory, Power BI, and key elements of Azure Synapse Analytics, with new functionality for data monitoring and observability, as well as an underlying shared foundation providing data security, governance and compliance. Microsoft Fabric is consumed as Software-as-a-Service (SaaS), with an interface called Workspaces providing users with access to the tools and functionality pertinent to their roles, as well as the ability to collaborate on analytics projects, along with generative AI-based tools such as Copilot.

There are seven Microsoft Fabric experiences, each providing functionality targeted at users in specific roles. Power BI in Fabric provides the company’s visualization and analytics functionality for business analysts and business users; Microsoft Fabric will also provide integration with its Copilot automated assistant, as well as its Azure OpenAI Service for generative AI. For data integration, Microsoft Fabric includes Data Factory, providing capabilities that Microsoft users will be familiar with from the Power Query data transformation and data preparation engine and Azure Data Factory serverless integration service. Data Factory provides connectivity to more than 170 data sources and more than 300 pre-configured data transformations for code-free extract, transform and load, as well as pipeline management functionality. Data Activator is a new data monitoring offering for business analysts, enabling them to monitor data environments and configure thresholds with functionality for alerting and automated actions.

The remaining four experiences provide functionality that is currently available via Azure Synapse Analytics, targeted at specific users based on their roles and responsibilities. Synapse Data Warehouse represents the next generation of data warehousing, while Synapse Data Engineering enables data professionals to create a lakehouse environment to collaborate on data integration, data warehousing, data science and BI projects as well as utilizing Apache Spark to transform data, with notebook-based development and monitoring. Synapse Data Science provides functionality for data preparation and code generation as well as ML model development and operationalization, and includes the SynapseML library for Apache Spark. Synapse Real-Time Analytics is optimized for data streaming and time-series workloads and provides automatic data streaming, indexing, and partitioning, along with auto-generated queries and visualizations.

The foundation of Microsoft Fabric is Microsoft OneLake. Based on Azure Data Lake Storage Gen2, Microsoft OneLake is an enterprise-wide data lake for storing files and structured or unstructured data. Cloud object storage is an increasingly critical component of organizations’ analytics strategies. More than one-half (53%) of participants in our Analytics and Data Benchmark Research already use object stores in analytics processes, while a further 18% plan to within 24 months. Microsoft OneLake is described as “OneDrive for Data,” providing a single environment for storage, management and governance that can be accessed by multiple analytic and integration engines. Microsoft OneLake also provides “shortcuts” to support virtualized access to data in external Azure Data Lake Storage Gen2 environments, as well as Amazon Web Services’ S3. Users can discover and explore data via the Microsoft OneLake data hub interface, which is integrated into the seven Microsoft Fabric experiences, as well as the Power BI desktop. Microsoft Fabric is also integrated with Microsoft Purview for data governance, risk, and compliance, and Fabric contains the Microsoft Purview hub page for administrators to manage and govern data in Fabric.

Most of the Microsoft Fabric experiences (apart from Power BI) are currently only available in preview, or in the case of Data Activator, listed as “coming soon.” As such, it remains to be seen how seamlessly Microsoft can stitch together the various experiences. It is also not clear at this stage if or how users of products such as Azure Synapse Analytics and Azure Data Factory will be able to migrate existing workloads to Microsoft Fabric. Nevertheless, I recommend that organizations exploring options for data management and analytics include Microsoft Fabric in their evaluations. The product is an ambitious addition to Microsoft’s product portfolio, providing SaaS-based access to functionality that has previously only been available via Platforms-as-a-Service offerings, with access tailored to the needs and responsibilities of users in specific roles.

Regards,

Matt Aslett