How to Query Database Backups in Microsoft Fabric Without Copies or ETL
In collaboration with Peleg Kazaz, and Liore Shai
Every company backs up its business-critical and operational databases. Every data-driven, AI-powered company also mirrors relevant data from its databases and consolidates it into a data warehouse for analytics and AI. These database backups contain the most complete, historical, and current view of the business, and are critical in a fast-moving, competitive market.
What if, instead of mirroring the data into the warehouse and consolidating it via complex ETL pipelines, backups could be queried directly, in place, and without moving or restoring them first?
Eon’s new Microsoft Fabric integration allows customers to unify their database backups stored in Azure, AWS, and Google into Microsoft OneLake, making them accessible from Fabric Data Warehouse, Fabric Spark, Azure AI Foundry, and many popular third-party tools.
Why Activate Your Backups
Backup data holds historical and operational context that can transform analytics and AI initiatives. Activating it unlocks new advantages:
- Query faster – Run analytics directly on backup data in place.
- Unify data – Unlock a centralized view across every data source.
- Spend less – Eliminate redundant storage, replication, and ETL pipelines.
- Do more – Use backups for compliance, disaster recovery, and cost visibility.
The result is a new approach where backups serve as a live, trusted data source for insight, rather than static storage for just-in-case scenarios.
Microsoft OneLake Shortcuts and Metadata Virtualization
Microsoft OneLake unifies all organizational data into a single, logical data lake. Through OneLake Shortcuts, customers can connect and access data wherever it resides, eliminating the need for duplication. Remotely referenced datasets are cataloged and made accessible via the OneLake Catalog, enabling engines to query the data as soon as it becomes available—using OneLake Table APIs, any Apache Iceberg-compatible engine can be utilized.
To unify access to data, OneLake builds on top of open table formats, such as Apache Iceberg and Delta Lake. To enable interoperability between engines that may support one or both of these formats, OneLake automatically and continuously converts between the table formats – this is a lightweight metadata-only operation and does not modify or move data.
This capability enables multi-cloud data access within seconds, allowing teams to work with consistent, governed data across environments and clouds.
Revolutionizing Cloud Backup with Eon
Eon takes a practical approach to cloud backups. It provides organizations with a single platform to manage their backups, choose where they’re stored, and access them instantly when needed.
Every data workload is stored in the open Apache Iceberg format. That means backups aren’t just for recovery—they’re structured, queryable datasets that can be used for analytics while costing less than traditional native backups.
In Microsoft Fabric, these Iceberg-based backups connect directly through OneLake Shortcuts, extending Eon’s secure storage model into a full analytics and AI environment without any extra data movement.
When combined with Microsoft Fabric, those capabilities extend beyond recovery or resilience—they become a foundation for analytics and AI.
What Eon Integration Adds for Fabric Customers
With OneLake Shortcuts, teams can easily connect Eon’s backups, hosted inside or outside Azure, and immediately leverage them in the Fabric ecosystem.
Eon publishes backup datasets as Iceberg tables and automatically connects them via OneLake Shortcuts. These datasets appear under the Tables directory in the OneLake catalog. This makes each Eon-managed backup instantly discoverable and queryable across Fabric workloads without modifying the underlying data.
Architecture: Eon’s Microsoft OneLake Iceberg Catalog Shortcut Integration
Business Impact: What This Enables
The following are a few ways organizations can use this integration to extract more value from existing data:
Analytics & BI at Scale
Utilize historical data from every system and cloud to drive trend analysis, forecasting, and informed decision-making in Power BI, without the need to restore or duplicate datasets.
Machine Learning and AI Readiness
Train and test predictive models on years of real operational data. Fabric’s integrated ML tools can now learn from full-context datasets managed in Eon.
Compliance and Risk Insights
Audit retention, residency, and policy adherence directly inside Fabric. Historical backups become searchable, governed sources for faster, automated compliance checks.
Cyber Resilience and Recovery Analytics
Go beyond recovery readiness. Analyze backup data to understand breach exposure, trace which systems or data were affected, and detect unusual access or change patterns.
Cost and Operational Optimization
Track storage growth, lifecycle, and redundancy to control costs. Identify unused datasets, automate tiering, and optimize retention based on real usage trends.
Proactive Monitoring and Alerts
Establish rules across databases to monitor key signals, including abnormal transaction rates, data drift, or policy violations. Get alerts directly from backup data for faster response.
The common thread: It’s no longer idle backup data. Now you can use it to fuel analytics, AI, and governance, activating years of context for smarter, faster business decisions.
How it Works
- Backups are written as Iceberg tables:
- Eon manages database backups and stores them as Iceberg tables, complete with metadata and data folders in Azure Blob Storage, Amazon S3, or Google Cloud Storage.
- Enable access in Eon
- Work with your Eon solutions architect or follow setup documentation to enable Fabric access for the selected backups in the Eon console.
- Verify configuration
- Choose automatic or on-demand access for the needed resources, then confirm Fabric has read-only connectivity to governed storage.
- Fabric virtualizes it
- The integration automatically creates a shortcut under the lakehouse tables directory, allowing Fabric to detect and register the table.
- The dataset appears in the OneLake Catalog
- It’s now queryable in Fabric Data Warehouse, Fabric Spark, Power BI, and any other engine integrated with OneLake.
Security and Governance
This integration inherits Fabric’s access and governance controls:
- Access remains read-only and is enforced at the dataset level.
- Authentication utilizes Microsoft Entra service principals, adhering to a least-privilege design.
- Shortcuts are continuously validated to ensure links point to governed, secure storage.
Administrators maintain full control while analysts gain governed visibility.
Getting Started
- In the Eon console, navigate to Integrations → Microsoft Fabric.
- Follow the Fabric connection guide to create a new connection.
- Configure access: choose Automatic Access (for continuous sync) or On-Demand Access (for specific tables or views).
- Verify that the connected dataset appears under the /Tables directory in Fabric.
- Connect Power BI or Fabric SQL to start querying.
What’s Next
Eon and Microsoft continue to expand interoperability, enabling any Fabric engine and popular third parties to access backup data as easily as production data. This collaboration allows customers to simplify data operations, enhance resilience, and lower cloud costs.
To learn more, request a demo or see it live at Microsoft Ignite 2025.
About Eon
Eon automates and activates cloud database backups, turning protected data into a queryable data lake for analytics, AI, and recovery. As the first platform built for Cloud Backup Posture Management (CBPM), Eon simplifies compliance, reduces cost, and powers insight across multi-cloud environments.
Through its partnership with Microsoft, Eon brings these capabilities into Microsoft Fabric’s OneLake, enabling enterprises to instantly access and analyze their backup data across Azure and other clouds to unlock new value for analytics, AI, and business intelligence. Learn more at Eon.io.
