Data modeling is the process of abstractly organizing all kinds of data in the real world, defining the scope of the database to be governed, determining the organizational form of the data, etc. until it is transformed into a real database. The data model is the core of building an application system, and it is a conceptual framework that represents business operations as accurately as possible.
The process of data modeling is the process of defining, analyzing, and discovering data requirements, and then representing such data requirements in a visual form ("model"). A data model is a set of symbols and words used to accurately represent communications in the information domain. A model of any landscape will include something (for example, a map is a model of a geographic landscape) and exclude something for ease of understanding.
"Discovery" is to determine what information the business needs in the business process or application, for example, understanding the customer and account number is an important concept. "Analysis" is the process of identifying requirements, such as a clear definition of customers and accounts over time, and an understanding of the relationship between customers and their accounts. "Scope" involves working with the business to decide what is most important for a particular business stage. For example, does the first stage require both "save" and "check account", or just "check account". "Representation" means showing what the information landscape looks like in clear and unambiguous language, such as the following data model:
-
A customer can have one or more accounts.
-
An account must be owned by one or more customers.
Once we have written these requirements into a data model document, we can communicate with the business and Information Technology (IT) personnel involved in application development, such as business users, business analysts, data modelers, data architects , database administrators, developers, testers, and administrators.
The data model is the primary medium used to communicate from business to IT, and within IT from system analysts, modelers, architects to database designers and developers. Whether the database technology to be used is a relational database management system (RDBMS) (such as ORACLE, Teradata), or a non-relational database (Not Only SQL, NoSQL) such as MongoDB or Hadoop, there needs to be a way Can be used to communicate data requirements. Therefore, we need data models.
The data model should be of high quality, able to support current needs while meeting future needs. The Data Model Scorecard is a tool that can be used to improve the quality of the data model.
Many of the clients I'm consulting with have decided to apply the data scorecard to the client's data model, and they also recommend using the data model scorecard to improve the design.
The big data employment challenge has a monthly salary of 30K. This article will briefly introduce the composition of the data model and teach you how to understand the data model.
1.1 Entities
An entity represents a collection of important and valuable transactional information related to a business. Each entity is represented by a noun or noun phrase and generally applies to one of the following 6 questions: who, what, when, where, why, how. Table 1-1 is the definition and example of these entity classes.