Verify the properties and relationships of the queried Project

1. Verify the properties of the query item

Object properties allow you to add additional information, such as descriptions and on-screen prompts. querying project properties allows you to edit and query project behaviors, such as the configuration data format. after importing the model, check the following two attributes:

(1) usage property)

Facts (data): Numeric (numeric value), time-interval (time interval), non-index column.

Identifiers: Key, index, date, time, and any index column.

Attributes (attribute): String, blobs.

Unkown (unknown): used when model design developers cannot determine the role of data

(2) regular aggregate Properties)

The general aggregation attribute of numeric data (numeric facts) describes how a method is aggregated. The default value is sum.

(3) another property set to be fully utilized is prompt Info (prompt information)

Prompt type: You can specify the type of prompt to be generated in studios or use the default prompt type. The default prompt type is determined by the server based on the data type.

Other attributes are used to improve the performance when the index is used for automatic retrieval, and the selected values can be displayed in a user-friendly manner.

Cascade on item reference (cascade of project references): used for Cascade prompts in RS.

Display item reference: used to identify the default value, that is, a manually created RS value is used to display a specific query item. Display the values of the current column to the values of other columns;

Use item reference: used to identify the default value, that is, the value used for the RS prompt created in the query filter. The currently displayed value must correspond to other values and use the corresponding values for operations.

Filter item reference: used to identify the default value, that is, a prompt generated by IBM Cognos uses this value to filter a query. The values displayed in QS must use the corresponding values of other columns in the query filter.

2. Check the relationship

The link is maintained in object digoal or context entity E. When verifying a link, you need to ensure that the appropriate link exists to meet your report needs, and you need to determine whether you need an optional or mandatory base. A selective base requires more processes, but it is likely to return the expected results.

Optional cardinality (selective base): an external connection is generated in SQL, which is expressed as 0, 0 .. n, or 0 .. 1.

Mandatory cardinality (mandatory base): an internal connection is generated in SQL, which is 1, 1. N, or 1. 1.

Cardinality (base): Used in IBM Cognos Bi to determine which query objects are facts (FACTS) and which are dimensions in the context of a query. This decision is important, especially when multiple fact tables are queried through the shared dimension. By identifying which query topics are facts, IBM Cognos Bi can properly aggregate these facts without losing any records of the fact table. The so-called base number is the record correspondence between the table and the table.

Modeling by star mode is very important because it ensures that the nature of a query topic is unambiguous. Simply put, fact query topics can only be appended with 1 .. N or 0 .. N base, while the dimension (dimension) query topic can only append 1 .. 1 or 0 .. 1 base.

 

3. For fact tables and dimension tables, first you need to understand the relationships and differences between the relational model and dimension model in the data warehouse.

Relational Models are mainly used in transactional databases. inmon is an advocate in the data warehouse field. It is a relational model based on 3nf. From the perspective of ERTU's "Perception, it is loose and fragmented, with a large number of physical tables and low data redundancy. Because the data is distributed in many tables, the data can be applied more flexibly and has powerful functionality. In the concept of inmon, DW (data warehouse) is not directly used for DSS/Bi and other applications. Instead, DW (data warehouse) is a platform with a 3nf relational model. For upper-layer applications, by creating a small data mart or other methods (dimension models can be used at this time) to meet specific application requirements-that is, the Data Warehouse is used as the data source of the data mart, data mart to meet the needs of specific applications (such as Bi and DSS.

Although the underlying layer of the dimension model is physically implemented as tables and relationships, the model concept has been abstracted. Tables are classified into dimension tables and fact tables. fact tables are dominated by numbers, A dimension table contains measurement information. dimension tables are usually text-oriented and often used as the "context" of a fact table to add business significance to those measurement values. The dimension model has a stronger business significance. Each table contains almost all the information of a topic, which conforms to the dimension (fact table) data of the same topic and is often integrated into a conform) in a dimension table (fact table), the number of tables is less than that of the relational model. Kimball is an advocate in the data warehouse field.