Getty Images/iStockphoto

Fannie Mae, Mercedes-Benz detail data lake platform uses

At the Subsurface Winter 2022 virtual conference, several organizations including Fannie Mae and Mercedes-Benz explained why and how they are using Dremio's data lake engine.

Large organizations are using data lake platforms more and more to serve data analytics and operational use applications.

At the Subsurface data lake virtual conference on March 2-3, large organizations including Fannie Mae and Mercedes-Benz outlined how they have moved to a data lake platform architecture.

The event was sponsored by data lakehouse platform vendor Dremio, which unveiled a series of new technologies including a query engine and a metadata store that use the open source Apache Iceberg technology. Iceberg provides a table format for data that is stored in a data lake.

In a breakout session on March 2, executives from mortgage financing organization Fannie Mae outlined how they use Dremio and a modern data lake architecture. Fannie Mae is one of the largest financial institutions in the world, managing over $4 trillion in assets, with regular reporting requirements to the Federal Housing Finance Agency.

"We were looking for ways to enhance and optimize our data flows, not only in finance, but across the enterprise as a whole," said Jeffrey Palmer, vice president of finance transformation, data strategy and program management at Fannie Mae. 

How Fannie Mae moved to a cloud data lake platform

Fannie Mae pulls data from many different sources, and a key challenge was making sure the data was accurate and authoritative, Palmer said.

As a strategic direction, Fannie Mae executives decided they needed a centralized cloud-based architecture that provided data governance, data lineage and the ability to power reports and analytics.

Screenshot of Mercedes Benz execs at the Subsurface virtual conference sponsored by data lake vendor Dremio.
Mercedes-Benz executives outlined how the carmaker is using Dremio's data lake platform to help unify the company's cloud data efforts.

In the same session, Kevin Bates, chief data officer at Fannie Mae, said moving to the cloud data lake platform focused on moving all the data sources, not just some of them, with a guiding principle of leaving no data behind.

We're trying to bring [data] together into a more elegant mechanism that's easier for us to manage and be accountable for.
Kevin BatesChief data officer, Fannie Mae

Fannie Mae's goal with the project was to create an authoritative data source for operational insights, Bates said.

"We're trying to bring [data] together into a more elegant mechanism that's easier for us to manage and be accountable for," he said.

Dremio's platform is now playing a key role in the Fannie Mae cloud data architecture. The agency conducted what Palmer called an exhaustive search to identify the right technology to provide a semantic data lake layer to organize and query data.

Palmer said Fannie Mae chose Dremio to enhance the agency's cloud data capabilities and to help get data to where it needed to go.

He added that the agency has linked the Dremio deployment it uses to multiple business intelligence tools that the organization also uses.

These include Tableau and MicroStrategy, which provide a self-service approach for business users to access data. He noted that before Dremio, different teams at Fannie Mae typically had to move data from one place to another in order to execute calculations. That's no longer the case.

"We have nearly completely eliminated all of the manual movement of data," Palmer said.

Driving cloud data forward at Mercedes-Benz

Automobile vendor Mercedes Benz has also moved to a cloud data lake architecture and is using Dremio to help with its operations.

In a conference session on March 2, Rodrigo Nunes, principal software engineer at Mercedes-Benz, said a key goal for his team with its cloud data platform was to make it easier for people in to the company to get better insights from data.

Mercedes-Benz vehicles, like those from most automakers, have multiple sensors in them that are always collecting data.

All that data has potential benefit and needs to be analyzed. The challenge was that the effort to effectively use of all the vehicle data wasn't unified within Mercedes-Benz, as different teams were building their own data pipelines to enable analysis.

Nunes said Mercedes-Benz decided to use Dremio to enable data lake operations in a unified approach, providing a platform that is used as the basis for data that any team at the company can use.

During the breakout session, Tyler Axdorff, staff cloud engineer at Mercedes-Benz, explained why the company chose Dremio.

"One of the key features that Dremio offered us that others didn't was the ability to join data across multiple cloud environments to be able to integrate with our catalog," Axdorff said. "So that was a big win for us."

Dig Deeper on Database management

Business Analytics
Content Management