Microsoft has launched a new feature called "Mirroring," introducing a significant capability for enterprise users of its Microsoft Fabric platform. This feature enables customers to replicate their external databases and manage them seamlessly within Microsoft’s data warehousing environment.
Mirroring stands out among the updates announced today as it simplifies the process of sharing data, even from proprietary formats. This enhancement not only consolidates different data sources within Microsoft Fabric but also helps organizations reduce costs.
The introduction of Mirroring intensifies the competition between Microsoft and major cloud rival Amazon, which has been slower to adopt open data formats. Industry analysts have commended Microsoft for its proactive approach since the announcement of Fabric in May. The pressure is also on Google, which is pursuing similar open format initiatives but faces challenges due to Microsoft's larger customer base and comprehensive service offerings. Additionally, Mirroring positions Microsoft against emerging analytics competitors such as Databricks and Snowflake, which have gained traction by facilitating data sharing.
This latest development reinforces Microsoft Fabric’s mission to enable companies to recognize, access, and analyze their data uniformly, regardless of its location. However, some analysts, like Tony Baer of dbInsight, suggest that such openness is becoming essential for cloud data providers, as many competitors are implementing solutions for migrating or replicating data from traditional on-premise systems. Baer notes, “To that extent, the Mirroring announcement is an example of keeping up with the Joneses.”
Other significant updates to Microsoft Fabric include:
- General availability of Fabric.
- Achievement of 25,000 customers since its preview launch in May.
- Public preview of Fabric’s Copilot feature, which allows users to create data flows, SQL statements, reports, and machine learning models using natural language.
- Expansion of Microsoft’s Independent Software Vendor (ISV) ecosystem.
Mirroring is designed to help organizations avoid costly SQL calls to external databases and eliminate complex migration projects—often referred to as the integration tax that arises during database integration following acquisitions.
One of the core components of Microsoft Fabric is OneLake, a unified data repository where customers can store all data in open formats like Apache Parquet and Delta. Mirroring facilitates the replication of cloud data warehouses and databases into these open formats, ensuring compatibility within Fabric’s environment, even if the original databases use proprietary formats. Starting today, this technology will support Azure Cosmos DB, Azure SQL DB, Snowflake, and MongoDB, with more data sources expected in 2024.
Arun Ulagaratchagan, Microsoft’s corporate VP of Azure Data, emphasized the importance of open data for driving business value, stating, “We really believe that all data needs to be open.”
Mirroring also extends the “Shortcuts” feature introduced in May, which enhances multicloud capabilities. Shortcuts enable Fabric’s OneLake to virtualize data stored in Amazon S3 and Google Cloud without data movement or replication. This feature is now generally available, though it is limited to databases in open formats like CSV and Parquet.
The London Stock Exchange (LSE), which manages one of the largest data repositories in the financial sector, is collaborating closely with Microsoft to test the Fabric platform and Mirroring. Dmitri Sedov, Global Group Head of Data Intelligence at LSE, noted that while the implementation is still in its early stages, early indications suggest that Microsoft’s vision for “Cloud 2.0” is on track.
Sedov highlights that Fabric shifts the focus away from data storage—an increasingly commoditized aspect of cloud services—allowing customers to concentrate on how they utilize their data. With features like Mirroring and Shortcuts, Microsoft Fabric enhances its appeal by enabling users to analyze and access data wherever it is located.
Despite being in the early phases of deployment, Microsoft reports strong interest in Fabric, with 67% of the Fortune 500 already utilizing the platform. Ulagaratchagan indicated that 85% of users are engaging with three or more workloads on Fabric, demonstrating significant growth.
He explained that Mirroring employs publicly available APIs from various database and data warehouse vendors to create and maintain near real-time snapshots of databases. This feature leverages change data capture techniques to ensure ongoing synchronization.
Once a source database is connected, users benefit from instant access to features like shortcuts, Direct Lake mode in Power BI, and a universal security model. Ulagaratchagan noted, “This provides massive performance acceleration and significant cost reduction.” For instance, when utilizing Microsoft’s Power BI, the application can directly access data in OneLake without executing SQL queries across mirrored databases.
Customers appreciate the integration of generative AI capabilities powered by Azure OpenAI, the streamlined data architecture that minimizes repeated data ingestion, and the unified business model that allows for a single purchase of Fabric to support all data workloads.
As Microsoft vies for prominence in the enterprise data landscape, the competition with Snowflake and Databricks is notable. Although Microsoft collaborates with Databricks through Azure Databricks, the introduction of Mirroring signals a strategic move to attract customers from these emerging data solutions, especially as some perceive the costs of Snowflake as excessive.
Hyoun Park, an analyst at Amalgam Insights, emphasizes Microsoft’s strategic goal of integrating as much data as possible into both Fabric and Azure, indicating that the company is preparing to capitalize on competitive perceptions while fostering innovation in the evolving AI landscape.
Ranking
