Data Lake

Problem

Your organization has: CSV JSON logs images events documents sensor data You don't yet know exactly how all of it will be analyzed.

Solution — Data Lake

Store large amounts of raw or semi-processed data in scalable storage. A data lake keeps diverse data for flexible exploration and processing; a data warehouse organizes curated data into tables optimized for analytical queries and reporting. For example, keep raw purchase events and product images in the lake, then load cleaned sales tables into the warehouse for monthly reports.

Core idea: a data lake keeps diverse data at scale before every future analytical structure is known.