Theories of Kimball and Inmon in relation to Datawarehouse design

Data warehousing is essential for Business developments. The corporate can improve the sales and profit using Datawarehouse. Many businesses have learned that they must either lead or survive — use their data wealth effectively. Due to the unique proposition as an integrated corporate data repository, the Datawarehouse performs a far more essential function in this circumstance. There are two architectural methods used today to create a data warehouse such as the Inmon method and Kimball’s method. Years of the debate have raged about which is significantly more accurate. However, no clear solution has ever been reached because both theories use their own advantages and differentiating factors.

Inmon’s Warehouse — (Top-down approach)

As per Inmon, a data warehouse is a centralized data storage facility. The “atomic” data is stored at the bottom of the hierarchy in a data warehouse. As per the Inmon Theory, dimensional data marts are created when a normalized data model is created.

The data warehouse is defined by Inmon as follows:

Subject-oriented: All the data components associated with a particular real-world event or item are grouped together in the data warehouse.

Time variant: Modification in the database’s data is tracked and documented, allowing for the creation of change reports that indicate changes with time.

Non-volatile: There will never be any overwriting or deletion of data in the data warehouse. The data will be static, read-only, and maintained for future reporting after it has been guaranteed.

Integrated: The database contains the data from the majority of the company’s departments, and it is all accurate.

Kimball’s Warehouse — (Bottom down approach)

A most crucial aspect of a business or field in mind, The first step is to set up a data mart. These provided access to corporate data and, if necessary, could be integrated into a bigger data warehouse.

Data is fetched from a variety of sources and loaded into a staging area using ETL technology. Data is loaded into a dimensional model here. The data warehousing dimensional model recommended by Kimball is not standardized. The star schema is the foundation of dimensional modeling.

Conclusion

Both Inmon and Kimball's approaches have proven to work for the successful delivery of data warehouses. We can not assume one is good other one is bad because Both have benefits and drawbacks, and they operate well in various scenarios. Normally, these approaches are practiced based on facts illustrated below.

The combination of both models is called a hybrid model. hybrid model, the data warehouse is developed using the Inmon model, and the business method Kimball method data marts are created using the star schema for reporting on top of the integrated data warehouse.

References

manmeetjuneja5, n.d. geeksforgeeks.. [Online]
Available at: https://www.geeksforgeeks.org/difference-between-kimball-and-inmon/

Sakthi Rangarajan, 2016.tdan [online]

Available at: https://tdan.com/data-warehouse-design-inmon-versus-kimball/20300

Sansu George, 2012. Computer weekly [Online]

Available at: https://www.computerweekly.com/tip/Inmon-or-Kimball-Which-approach-is-suitable-for-your-data-warehouse