• Source -> Staging -> Datawarehouse -> Cube -> Reports,


    Source data, target data, data warehouse, data mart, reports, ETL - Extract Transform Load, etl logic - incremental or full , integration services, reporting services, analysis services, database engine, cubes, measures, measure groups,dimensions, facts, reports testing - drill down, drill through, build deployment. I mean there are just so many of them .
    How does the chain look like ? May be something like this :
    Source  -> Staging -> Datawarehouse -> Cube -> Reports, Oracle, MySql, SAP, Teradata, DB2, Sybase. It can be any combination of the listed as well as non listed ones for example as simple as flat text files. These multiple data sources can be an input data source to an application software that might help the end user take future strategic decisions.
    Staging : An intermediary state of source data that act as an input to the datawarehouse where in data is loaded by the ETL logic using integration services from the source data. It is more or less the source data in its original form with those tables getting discarded that might not be needed for the report generation within the application.
    Datawarehouse : It is the final entity developed within the database engine that gets created in the sequence mentioned above. It is here that the final objects get created and based on these objects the cubes are created for the analysis purpose. Here the object types that the complete data is organized into is called dimensions and facts. They as well as just like simple tables but the attributes within each dimensions and facts have a specific association within itself to resemble some realistic facts and associated information as per the business requirement.
    Cube :  This is the Analysis services objects within the complete  BI application development phase. As the name suggests it is not just the two dimension tables that hold its relevance in a typical RDBMS . It is indeed a three dimensional data modeling technique wherein we analyze a data set in more than just two dimensions. The facts that are associated with an application are analyzed as per the association they have with the various parameters which in BI terms are called as measures, or measure groups. Measure groups are actually a combination of more than one related measure. Thus we get greater insights into how a specific aspect of any business decision making gets impacted with a variation in various parameters.
    Reports : Reports are nothing but the cubes data getting represented onto a user friendly interface with the option to parameterise the reports as per the business needs. In simple terms it is the end product that gets developed and the data we look for in the cubes are available for view purpose in them. We can drill down and drill through them based on the scenario we need. For example we can by default see the report for any specific fiscal year as to how much sales have been materialized and then drill down onto the quarter basis , and then monthly basis, then the weekly basis and finally on the daily basis . Similarly drill through also gets applied on to the reports and the data can be visualized as per needs. Authorization and authentication is another feature that has its role to be played in the reporting services but then the authorization of the cubes over ride the privileges granted on the reports.

  • 相关阅读:
    阿里云标准-Docker安全基线检查
    阿里云标准-Redis安全基线检查
    阿里云标准-Nginx安全基线检查
    python中urllib.request和requests的使用和区别
    python爬百度首页
    进行base64编码
    python基础入门 文件操作
    python面向对象编程: 搬家具案例
    在vi或vim上查找字符串
    浅析数据一致性
  • 原文地址:https://www.cnblogs.com/stay-sober/p/4390790.html
Copyright © 2020-2023  润新知