0

all

I am recently pondering that how to classify data into different categories in Erp solutions, basing on that, I can decide which data should I strip out and put it into a shared database for multiple tenants instances.

As a industry practice, the Erp product is separated into 2 layers. The technology platform layer provide a lot of reusable components and modeling tools, make business applications follow the consistent architecture, the business application layer witch based on it provide the business functions.

So, basically the data can be categorized into 2 main types.one is platform data, the other is business data.Further，the platform data can be categorized into sub categories:

  1. platform 1)environment 2)engine related(Form engine, workflow engine, data access engine...which make the business function work) 3)metadata(for example:Form Description,Business Object Description, Data Model, Workflow definition) 4)configurations(organization or user related configurations) 5)management related(data structures for manage the models)
  2. business 1)model instances(actual orders data) 2)business configurations 3)derived data(from model instance data, and form query or analysis)

After analysis, I found that the environment data, configurations, management related data, business data are in a high degree of coupling. The only category can be separated from the instance database is the metadata.

1.Does my analysis reasonable? 2.Are there any patterns for reference?

Thanks.

zhengwb
  • 1
  • 2

1 Answers1

0

I would like to suggest the following pattern for splitting the data in the database

  1. MetaData[3] that you use to identify, authenticate and authorize the users, the Settings or customization Congifurations [4] per users or tenants along with the basic and the other components [1 throught 4] that are platform specific, (because you have not specified what environment does etc..) and common for all tenants can be in a separate database and the rest of the business specific data to reside in another database[s].

This will help the tenants that come from a geographically different location to store their own data in their separate database, even as part of the national laws and regulations and data safety.

Post your understanding in this regard

Saravanan
  • 7,637
  • 5
  • 41
  • 72