Premium Video

This video is available to premium subscribers only

Unlock Full Access

Many companies keep their analytical data in two places: a data lake for cheap, flexible storage and a data warehouse for reliable SQL reporting. Keeping the two in sync costs money and effort, and the copies drift apart over time.

A data lakehouse tries to give you both in one system. The data stays in open files on cheap object storage, and a table format such as Apache Iceberg, Delta Lake, or Apache Hudi adds database features on top: transactions, updates and deletes, time travel, and schema checks.

In this chapter, we will look at why lakehouses exist, how open table formats turn plain files into reliable tables, and the trade-offs of using one platform for both analytics and machine learning.