Introduction to data warehouse layering (real-time data warehouse architecture)

Hello everyone, meet again, I'm your friend Quanstack Jun.

Introduction to Data Warehouse Hierarchy

1. Advantages of data warehouse layering: simplifying complex problems; reducing repetitive development; isolating original data.

2. The concrete realization of data warehouse layering

ODS (Operation Data Store) layer: the original data layer, save the original data, directly load the original log, data

DWD (Data Warehouse Detail) layer: The detailed data layer is also called the DWI layer. The structure and granularity are consistent with the original table, and the ODS layer data is cleaned (removing null values, dirty data, data exceeding the limit range, row formatStore to columnar storage, change compression format)

DWS (Data Warehouse Service) layer: the service data layer, which is lightly aggregated based on DWD.For example: the day of the user, the day of the device, the day of the product.

ADS (Application Data Store) layer: The data application layer is also called the App layer or the DM layer, providing data for various statistical reports

Publisher: Full stack programmer, please indicate the source: https://javaforall.cn/126974.htmlOriginal link: https://javaforall.cn