Combining all your data in a traditional data warehouse is an anti-pattern, which requires a lot of ETLs on a rigid platform that mainly supports structured data geared toward mission-critical reporting and BI use cases. To leverage ML and AI effectively and be a better, more data-driven organization, you need a unified repository that supports large and diverse data sets, including semi-structured and unstructured, in an open format with enterprise-grade reliability and performance. Welcome to lakehouse, a new architecture that combines the best elements of data lakes and data warehouses. This packaged offering from Databricks will accelerate implementing one simplified platform for data analytics, data science and ML — setting the foundation for your lakehouse vision.