- Paradigm: Normalization rules for relational models .
Three database paradigms proposed by Codd:
1. First Paradigm
The first paradigm requires that each row in the table must be unique. Because relational databases are based on set theory, the definition of a collection requires that each element is unique (it can be distinguished by a key in a relational database). Therefore, all relational database tables must satisfy the first paradigm.
2. Second Paradigm
The second paradigm requires that the first paradigm must be satisfied, followed by the requirement that a certain condition must be met between the non-key attribute and the candidate key value attribute. That is, all non-key value properties are completely dependent on all attributes with the entire candidate key value.
How to understand noun reference http://www.cnblogs.com/lumnm/archive/2010/11/08/1871783.html
3. The third paradigm
The third paradigm requires that the second paradigm be satisfied, followed by non-key-value attributes that cannot be dependent on other non-key value attributes.
Summary: After actually satisfying the third paradigm, each table, that is, each relationship generates a relationship element. This relationship cannot be detached from other relationships.
- The life cycle of the data
Oltp-->dw-->olap
First, access to the online transaction processing system, mainly related to the increase and deletion. Then into the data warehouse, for the export with the data report, and then into the online online analysis system, for data query.
A database engine can install multiple DB instances, but there is no relationship between the two instances. But will share any computer resources such as CPU, memory, etc.
Database
Databases can be understood as containers for various objects, tables, views, stored procedures, etc.
The database system includes:
1.master Save meta-information of SQL instance, server configuration, etc.
2.Resource actually holds all the database objects
3.model is actually a template for the database
4.TEMPDB where temporary data is saved
SQL Technology Insider One