Data fabric, another step in evolving data architecture, uses metadata, machine learning, and automation to provide a unified view of enterprise data regardless of its format or location. It encompasses previous architectures—data warehouses and data lakes—to support business intelligence, data science, and embedded analytics.