Launch Deal · 50% off 1st month ·

2.4 — Create Shortcuts and Implement Mirroring

DP-700 Domain 2 · Skill 2.4

What You Need to Know

Shortcuts and mirroring both bring outside data into OneLake without building an ETL process. A shortcut references data where it already lives, so nothing is copied. Mirroring continuously replicates a database into OneLake as Delta tables.

Shortcuts

Fact Detail
What it is A pointer that appears as a folder. Deleting the shortcut doesn't delete the target, but deleting files inside a shortcut does delete them in the target if you have write access there
Where you create one In a lakehouse (under Tables at the top level only, or anywhere under Files) or in a KQL database (queried with external_table())
Internal targets Other lakehouses, warehouses, KQL databases, SQL databases, mirrored databases and semantic models, in any workspace
External targets ADLS Gen2, Azure Blob, Amazon S3 and S3-compatible storage, Google Cloud Storage, Dataverse, OneDrive and SharePoint, Iceberg tables. On-premises sources work through a data gateway
Tables A Tables shortcut to a Delta folder is recognized as a table automatically, and schema changes in the source sync on their own
Security Internal shortcuts use the caller's identity, so the reader needs permission on the target. External shortcuts authenticate with a stored cloud connection
Cost control The workspace shortcut cache (S3, GCS, S3-compatible, gateway) cuts repeated cross-cloud egress charges
Limits Up to 100,000 shortcuts per item, 10 per OneLake path, and 5 shortcuts chained to each other

Use a shortcut when the requirement says "don't copy the data", "avoid duplicating it" or "query it in place", and the source is file or table storage.

Mirroring

Type What happens Sources
Database mirroring Changes replicate continuously (as often as every 15 seconds) into read-only Delta tables, with a SQL analytics endpoint Azure SQL Database and Managed Instance, SQL Server (through a gateway), Azure Cosmos DB, Azure Database for PostgreSQL, Snowflake, Oracle, Google BigQuery, SAP and others
Metadata mirroring Syncs only the catalog (schemas and tables). The data stays at the source and is read through shortcuts Azure Databricks Unity Catalog, Snowflake
Open mirroring Your own application writes change files into a landing zone, and Fabric merges them Any system, through the public API
  • Mirroring is near real time and fully managed. You don't schedule it or write pipelines for it.
  • The Fabric compute that replicates the data is free, and mirrored storage is free up to a limit tied to the size of your capacity.
  • Mirrored tables are read-only in Fabric. To transform them, read them with Spark or T-SQL and write somewhere else.

OneLake availability for eventhouses

An eventhouse doesn't write its data to OneLake by default. Turning on OneLake availability makes a read-only Delta copy that other engines (Spark, the SQL endpoint, Direct Lake) can query.

  • Database-level: new tables and their data are made available. Choose Apply to existing tables to backfill tables that already exist.
  • Table-level: makes that table and all its data available.
  • The Delta files are written in batches, which can take up to 3 hours by default. TargetLatencyInMinutes can lower that to as little as 5 minutes, at the cost of more small files.
  • While it's on, you can't rename tables, change column types, apply RLS, or delete or purge data.

Key Concept: A shortcut is for files and tables you can point at. Mirroring is for operational databases you want replicated continuously. Copying with a pipeline is for when you must transform the data or keep your own snapshot.

Exam Traps

  • Copying S3 or ADLS data into the lakehouse with a pipeline when the requirement says not to duplicate it
  • Creating a Tables shortcut inside a subfolder of Tables (only the top level is allowed)
  • Expecting a mirrored database to accept writes or transformations
  • Expecting an eventhouse's existing tables in OneLake without choosing to apply availability to them
  • Forgetting that reading an internal shortcut needs permission on the target item

Check your understanding

Practice Scenario

🔒 This practice scenario is part of Skillmarked Pro

The lessons stay free. Pro unlocks the other 16 DP-700 practice scenarios and the 50-question timed mock exam, plus everything for every other exam.

See plans

Launch Deal: first month €9.95, then €19.90/month. Enter code LAUNCH50 at checkout. Ends 1 December.

Cancel anytime. Or €199 a year, which saves 17%.

Not sure yet? Try 10 free mock exam questions.

Already Pro? Log in.

Sources