SQL Server2005 Analysis Service Practice starts

Defining a data source view in an Analysis Services project

1. Create an Analysis Services project based on the template

BIDS (Business Intelligence Development Studio) uses templates to create different types of projects, and the Analysis Services project is one of the templates, and the templates are customizable.

2. Define the data source

Connect SQL Server using the native OLE db\microsoft OLE DB Provider for SQL Server driver.

3. Define the data source view

Data source view, which is the metadata view of the table or view of the data source selected in the project, stores the metadata in the data source view, allowing the user to detach from the data source during the development process using meta data. You can select multiple heterogeneous queries as data sources, but at least one data source must be a SQL Server data source.

• Commodity dimension: Pub_ Commodity main file
• Date Dimension: Sys_ Sales Calendar
• Store Dimension: Pub_ Warehouse Store
• Vendor Dimension: pub_ Suppliers
• Fact: E1 Day sales details

4, modify the default table name

Bids defines dimensions (dimensions), attributes (attributes), and measure groups (measure groups) using metadata in the data source view. Bids use the FriendlyName property of a cube object instead of the Name property, this step is primarily to modify the FriendlyName properties of the object created in the previous step to improve user-friendliness. (Not only can you change the friendly name of the table in the data source view, you can also change the name of the column, define computed columns, and even connect between tables or views to improve user friendliness).

Several concepts:

• Dimensions (Dimension): The basic component of multidimensional cubes (a fundamental component), which organizes data (in a logical hierarchical format to organize business data) from the perspective of user interest.
• Attribute: A component of a dimension (building block) used to organize measures in a multidimensional cube. A dimension is a collection of attributes that correspond to one or more columns in a dimension table. In a dimension, attributes are organized in a hierarchical manner, providing the path of aggregation drill-down for the measure in the cube. property to provide a grouped, constrained information to the client application.
• Measure (Measure): A column in a fact table (fact table) that contains numeric data that can be calculated, aggregated, and so on. In a cube, measures are grouped by fact tables into measure groups.
• Measure group (measure groups): Used to group measures contained in a fact table and to communicate between dimensions and measures.

• Pub_ Commodity main file-->product
• Sys_ Sales Calendar-->date
• Pub_ Warehouse Store-->depository
• Pub_ Suppliers-->supplier
• E1 Daily Sales Detail-->sales