Federation and the Lakehouse: Two Roads to Unified Data Access, and How to Know Which One to Take

This article provides a comprehensive analysis of two approaches to unified data access in modern data architecture: federation (querying data in place) and the lakehouse (consolidating data into a governed store). Written by Alex Merced of Dremio, it covers the mechanics, strengths, and limits of each, including failure modes. A decision framework helps choose the right approach for specific workloads, and a lifecycle pattern of 'federate first, promote deliberately' is presented. The article also discusses best practices, anti-patterns, economics, and a worked example, emphasizing that the two roads are complementary rather than rivals, especially with the rise of AI agents.