[Data warehouse design] Why should enterprise data warehouses be layered?(six benefits)

I. Should the data warehouse be layered?

Since the data warehouse is the place where data is stored and calculated, why do you need tiering?It is also the data scale, which is determined by the business scenario.It can be said that when the data warehouse construction of many companies is just starting, most of the data is subjected to rough data access. After ETL is performed, it is directly connected to the business, and reports are generated or imported into the business system for direct use.

Later, with the development of the company's business, the precipitation of data, and the development of the data warehouse to a certain stage, it was found that the use of data was disorganized and various businesses were directly calculated from the original data.This results in various repeated calculations (maybe the two tables are only a few fields apart, but everyone ran it once), which seriously wastes computing resources and storage resources, and costs enterprises a great deal.At this time, everyone has to think about how to standardize storage and computing, and how to minimize enterprise costs.In particular, companies with larger data scales have stronger demands.
Of course, your company's data scale is small, and it is not necessary to have no layering. Of course you can.There is no need to implement such norms. The disadvantage of norms is that they have to pay a lot of labor costs to implement norms and supervise the implementation of norms.The final choice should be considered in combination with the cost of your business, and everything should be based on reality.

Second, the benefits of data warehouse layering

1. Clear data structure:

Each data layer has its scope, so that we can more easily locate and understand when working with tables.

2. Data bloodline tracking:

In simple terms, it can be understood in this way. What we finally give to the business integrity is a business table that can be used directly, but there are many sources of it. If there is a problem with a source table, we hope to be able to quickly and accurately locate it.problem and understand the scope of its harm.

3. Reduce repetitive development: