Embedded data marts for central data warehouse

摘要

A central data warehouse includes embedded data marts. These embedded data marts, referred to as workspaces, are assigned centrally manage data by reference only but rely directly on the centrally managed data and the underlying infrastructure. Workspaces still allow departments in an enterprise to perform certain actions on their own (like adding new data and building new models) without having to instantiate copies of the centrally managed data in a locally managed data mart.

主权项

1. A method in a computer system comprising a processor and a memory, the computer system hosting a data warehouse, the method comprising operating the computer system to perform steps of:
receiving source data from a plurality of data sources in an enterprise; storing the source data, as centrally managed data, in a data storage system configured with a plurality of data tables that comprise the data warehouse, including:
using an extraction, translation, and load (ETL) tool to process the source data and storing the processed source data in the data warehouse;using the processed source data to generate a plurality of enterprise data models and storing the enterprise data models in the data warehouse; andusing the enterprise data models stored in the data warehouse to generate a plurality of enterprise data views and storing the enterprise data views in the data warehouse; creating a workspace, including:
assigning at least one of the enterprise data models and enterprise data views stored in the data warehouse as referenced data to the workspace by reference only; andassigning a plurality of first users to the workspace; granting only the first users read and write access to the workspace, the first users having read-only access to data that correspond to the referenced data; accepting only from the first users local data that is external to the computer system; and defining a user data model in the workspace using the referenced data and the local data.