Azure Data Lake

The external sources are the exciting part.

The list reaches Azure Data Lake Storage Gen2, Azure Blob Storage, Amazon S3 and S3-compatible storage, Google Cloud Storage, Dataverse, OneDrive, and SharePoint, and even on-premises or network-restricted locations through a gateway.

There's also support for iceberg tables, which we'll come back to in a moment.

Read that list again and notice what it means.

Spark notebooks can read it, SQL can query it when it's a table, and the real-time tools can use it too.

Power BI can read it through a semantic model, including indirect lake mode, which we'll come back to later.

Even applications outside Fabric can reach it through the OneLake API, because OneLake speaks the same storage language as Azure Data Lake Storage.

Iceberg deserves a plain explanation, because it's a name you'll hear often.

The trade-off is less flexibility, but the payoff is speed.

Synapse takes a more modular Azure approach, giving you SQL pools for data warehousing, serverless SQL for querying files, Spark for large-scale processing, pipelines for moving data, and Azure Data Lake storage for the file layer underneath.

Those pieces can work together, but your team decides how to configure, secure, scale, and connect them.

The trade-off here is more control, but also more complexity.

Teams store and access data using open formats like Delta and Parquet.

A lakehouse, a warehouse, and a Power BI semantic model can all draw from the same foundation instead of creating private copies.

And OneLake supports shortcuts, which let Fabric reference data already living elsewhere like Azure Data Lake Storage without forcing you to copy files first.

That matters when your company has years of data in Azure Storage and doesn't want a major move before it can start using Fabric.

The practical shape of Fabric matters more than the feature list.

OneLake sits on Azure Data Lake storage, every workspace is a folder inside it, and you don't even need an Azure account to use it.

You buy Capacity, which is a pool of computing power sized by a level Microsoft calls a SQ, and Azure builds it per second with per-user licenses for the people doing the work.

Then there's Azure Synapse Analytics.

If the work is large-scale engineering, machine learning, or streaming data, Databricks gives you the most control over the engine and the compute.

The more useful answer is that you often don't have to choose at all.

Microsoft documents shortcuts in one like that point at data living elsewhere, including Azure Data Lake Storage, Amazon S3, and Google Cloud Storage, with no copying and no migration.

A shortcut can even point at a mirrored Azure Databricks catalog, and if those files are in delta format, the lakehouse recognizes them as tables and keeps the schema in sync on its own.