iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
Microsoft connects enterprise data through several complementary Fabric options, not one universal connector. OneLake provides a shared namespace for Fabric workloads; shortcuts can reference supported data without automatically making another copy, mirroring integrates external databases or catalogs with source-dependent behavior, and pipelines or other ingestion tools move data when that better fits the need. The right choice depends on the source, access requirements, latency, transformation, and governance.
How the architecture fits together
Think of the design as a sequence: connect to a source, make its data available to Fabric, organize or transform it as needed, and govern how people and workloads use it. OneLake is the common data layer in that picture, but it does not make every source behave identically. The integration method, source capabilities, workload, and identity configuration all matter.
Connect the source
Fabric Data Factory connectors and ingestion options can connect cloud and on-premises systems, databases, SaaS applications, files, and event sources. The available connector and its prerequisites depend on the particular source. Fabric also provides pipelines with Copy activities, Copy job, and Eventstreams for different movement and ingestion needs.
Expose or move the data
A shortcut references selected data at another supported location so Fabric workloads can access it through OneLake’s namespace. It does not, by itself, mean that Fabric creates a second copy of the referenced data. Mirroring is a separate integration pattern for an external database or catalog; depending on the source, it may access data in place or replicate it. If data must be brought into Fabric, or transformed during transfer, use an appropriate ingestion or copy path.
#1 Best Overall
Curate and consume
One documented Microsoft reference architecture arranges data into bronze, silver, and gold layers: retain raw source data, conform and reuse datasets, then publish curated models for analytics. This is an example, not a mandatory Fabric layout. Governed outputs in that architecture support Power BI, data agents, Copilot, and operational reporting. A separate pattern uses Dataverse virtual tables to expose Fabric lakehouse data to Power Platform apps and flows.
Choose the connection method for the job
| Option | What it does | When it may fit | Check before choosing |
|---|---|---|---|
| OneLake shortcut | References selected data in a supported internal or external location. | You want Fabric workloads to access supported data without an initial copy, such as selected open-format data. | Source and format support, credentials, permissions, caching, identity mode, workload support, and what happens if the target moves or is deleted. |
| Mirroring | Adds an external database or catalog to Fabric; source-specific behavior may be in-place access or replication. | You need database- or catalog-level integration and the source is supported. | Whether the source is replicated or accessed in place, latency, supported objects, and resulting storage and operational behavior. |
| Pipelines, Copy, or Data Factory | Moves data into Fabric and may transform it. | You need managed movement or transformation, or a source is not suitable for a shortcut. | Connector support, refresh and latency needs, transformation, residency, and ongoing pipeline operations. |
| Power Apps Link to Microsoft Fabric | Makes Dataverse data available in OneLake through shortcuts. | You are bringing Power Apps or Dynamics 365 data into Fabric analytics. | The documented Dataverse shortcuts are read-only, so this is not an application write-back path. |
Shortcuts and mirroring are not mutually exclusive. A design can use a shortcut for selected files or tables and mirroring for a supported database or catalog, then use pipelines for data that needs movement or transformation. Microsoft describes shortcuts as operating at selected table, folder, or file granularity, while mirroring can support open and proprietary formats. Verify support for the exact source and workload before designing around either option.
Rank #2
Dataverse has a direct analytics path
Power Apps Link to Microsoft Fabric makes Dynamics 365 and Power Apps data available in OneLake through shortcuts while the source data remains in Dataverse. Microsoft describes this pattern as avoiding export and ETL construction for that direct link. The shortcut is read-only and uses delegated authorization with the credential specified for the shortcut, according to Microsoft’s setup documentation. Treat it as an analytics access path, not as a route for an app to write changes back to Dataverse.
This is distinct from exposing Fabric lakehouse data to Power Platform: the documented virtual-table pattern goes in the other direction, making Fabric data available to apps and flows. Choose based on which system is the source and which side needs to consume the data.
Plan identity and governance across the design
Access control is not a final layer to bolt on after integration. Microsoft’s reference architecture applies identity, role-based access control (RBAC), lineage, deployment controls, and certified semantic models across ingestion, data layers, and consumption. Permissions and identity behavior should be planned for each source, shortcut, Fabric workload, and consuming user or service.
- Validate the identity path. Do not assume a shortcut always uses the end user’s identity. OneLake documents identity-mode limitations, and Dataverse shortcut setup specifies delegated authorization with the credential selected for that shortcut.
- Map permissions end to end. Account for access at the source, the OneLake or Fabric layer, and the consuming workload; a shared namespace does not make source permissions irrelevant.
- Track change and lineage. Include lineage and deployment practices in the design so teams can understand how data moves or is referenced and manage changes to the solution.
- Govern published models. Use governed, certified semantic models where appropriate for analytics consumers rather than treating raw access as the finished product.
A practical way to decide
- Identify the source and required granularity. Establish whether the source is a file or open-format dataset, a database or catalog, Dataverse, or an event stream; then confirm current support and prerequisites for that exact source.
- Decide whether data should remain in place. If referencing supported data meets the need, assess a shortcut. If database-level integration is required, assess mirroring and determine whether that source is accessed in place or replicated.
- Choose movement when it serves the requirement. Use connectors, pipelines, Copy activities, Copy job, Eventstreams, or another suitable ingestion path when data needs to be moved, transformed, refreshed on a defined schedule, or handled by a source-specific connector.
- Test operational behavior before rollout. Confirm credentials, permissions, workload compatibility, latency, caching, residency, and recovery implications. For shortcuts, also consider changes to the referenced target; for mirroring, confirm supported objects and storage behavior.
- Design curation and controls with consumption. Decide how data will be organized and transformed, who can access it, how lineage and deployment are managed, and which governed outputs serve reports, agents, apps, or flows.
Source support, authentication behavior, and feature limitations can change. Check Microsoft’s current documentation for the specific source and Fabric workload before committing to an implementation.
Quick Recap
Best Value
Rank #4
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →
