According to Inmon (2002) quoted by Niu (2009), in general business intelligence system architecture have consists of four levels of components and meta data management module. The components interact to facilitate basic functions of business intelligence: extract data the companys operating system, save the data that has been extracted into the data warehouse and interesting data stored for a variety of business analysis applications.
1. Operating system level.
As the data source of business intelligence systems, business operating systems generally use a system of online transaction processing (OLTP) to support daily business activities, in general OLTP system is customer order entry systems, financial systems, and human resource systems. Operating system level.
2. Level data acquisition.
At this level is a pre-process component consists of three stages: extraction, transformation, and insert (ETL). A company has several OLTP systems that generate a very large amount of data. The data are first extracted from the OLTP system by ETL processes and then transformed according to the transformation rules. If the data has been transformed, the data are entered into the data warehouse. ETL is a basic component of business intelligence systems for quality data from other components depending on the ETL process. In the design and development of ETL, data quality, system flexibility and speed of the process is a major concern.
3. Level data storage.
Data that has been processed by ETL components are stored in a data warehouse, which is usually implemented using traditional database management system (RDMS). RDMS is designed to support the transaction process, in contrast to the focus on the subject of data warehouse, a variant of time and stored in an integrated way. Star schema and snowflake schema data warehouse is the most popular. Whatever scheme is used, the type of tables in a data warehouse is the fact tables and dimension tables.
4. Analytical level.
Based on the data warehouse, a wide range of analytical applications have been developed. Business intelligence system supports two basic types in analytical functions: reporting and online analytical processing (OLAP). Reporting function provides managers of various types of business reports such as sales reports, product reports, and reports of human resources. Reports generated from running queries into the data warehouse. Data warehouse queries in general has been defined by the data warehouse developers. Reports generated by the business intelligence system usually has a static format and contains the exact data type. See also Business intelligence report or 7 advantages of business intelligence.
1. Operating system level.
As the data source of business intelligence systems, business operating systems generally use a system of online transaction processing (OLTP) to support daily business activities, in general OLTP system is customer order entry systems, financial systems, and human resource systems. Operating system level.
2. Level data acquisition.
At this level is a pre-process component consists of three stages: extraction, transformation, and insert (ETL). A company has several OLTP systems that generate a very large amount of data. The data are first extracted from the OLTP system by ETL processes and then transformed according to the transformation rules. If the data has been transformed, the data are entered into the data warehouse. ETL is a basic component of business intelligence systems for quality data from other components depending on the ETL process. In the design and development of ETL, data quality, system flexibility and speed of the process is a major concern.
3. Level data storage.
Data that has been processed by ETL components are stored in a data warehouse, which is usually implemented using traditional database management system (RDMS). RDMS is designed to support the transaction process, in contrast to the focus on the subject of data warehouse, a variant of time and stored in an integrated way. Star schema and snowflake schema data warehouse is the most popular. Whatever scheme is used, the type of tables in a data warehouse is the fact tables and dimension tables.
4. Analytical level.
Based on the data warehouse, a wide range of analytical applications have been developed. Business intelligence system supports two basic types in analytical functions: reporting and online analytical processing (OLAP). Reporting function provides managers of various types of business reports such as sales reports, product reports, and reports of human resources. Reports generated from running queries into the data warehouse. Data warehouse queries in general has been defined by the data warehouse developers. Reports generated by the business intelligence system usually has a static format and contains the exact data type. See also Business intelligence report or 7 advantages of business intelligence.
0 komentar:
Posting Komentar