Data Lake and Data Lakehouse


Data lake is a technology for storing large amounts of structured and unstructured types of information in their original format.

  • used to store data that is collected but not yet ready for cleansing and analysis
  • allow organizations to gather structured, unstructured, and semi-structured data in large quantities
  • serve as a kind of catchall for data
  • not really meant to be more than a super-flexible storage place
    • simply allows you to store data when you don’t have a structured place for it
  • provide cost effective storage
    • because not paying the same costs that are tied to structured data
  • don’t require pre-designed schemas
    • perform a function called schema-on-read
      • builds the schema for the data as it goes into the data lake
    • differs from data warehouses and data marts, which use schema-on-write
      • data flows into the pre-designed schema

Data Lakehouse

Data lakehouse is a data management system that combines the best of both data warehousing and data lakes.

  • provide flexibility, like a data lake
  • more cost effective than a data warehouse
  • serve up information for data analysts, data scientists, and data engineers
  • support business intelligence and analytics projects in addition to machine learning and data science