Data Warehouse and enterprise application integration (i)

Data
Main content

A. Considering the development of data warehouse from the perspective of system and whole
Two. Concept and content of CIF
Three CIF Case-sap BW
Four Data Warehouse and Enterprise application integration
Five Summary

References

Summary

The main content of this paper is to introduce the enterprise information factory and discuss the system and integrity that should be considered when establishing data warehouse and enterprise application.

Therefore, the latest development of the main needle data warehouse, combined with the actual case of SAP BW, this article tells the Enterprise Information Factory (CIF) The concept, concept and content, but also elaborated in the design enterprise application to take the thinking way, we will look at the data in a new perspective.



I. Consider the development of data warehouse from the perspective of system and whole.

Given the complexity and scale of the problems that people are facing, people often tear the world apart and process it, which obviously makes complex problems easier to deal with, but in the end we pay the price--and we don't understand the series of consequences of our actions. This error is often repeated in the development of enterprise applications, and is no exception in data warehousing, and we need a systematic and holistic view to re-examine and reflect on the development of data warehouses and their relationships with enterprise applications.

Data warehousing is a theme-oriented, integrated, non-renewable (Stable), data collection that changes over time (at different times) to support the decision making process in business management-W.h.inmon



The essence of Data Warehouse is to use data support decision, its core value is whether the data model portrays the core business of the enterprise, there is a view that the collection of data mart is the Data Warehouse, the summary record can be put into Data warehouse or data mart supported by Data Warehouse. When the summary record enters the data warehouse, it is oriented to the general application, and the Data mart is adapted to the department application. The most obvious solution is usually ineffective; there may be some improvement in the short term, which will only worsen things in the long run.

Enterprise's activity, also is a kind of "system", also is affected by the subtle and closely related actions, each other influence, this kind of influence often has to be full of over the period of the period to show out. Being a small part of the community and wanting to see the whole change is doubly difficult. We tend to focus on a fragment of the system, but we can't figure out why some of the most fundamental problems seem to have never been solved.





Two. Concept and content of CIF

CIF (Enterprise Information Factory) is a systematic view of the architecture of the Data Warehouse system, we can change the angle to look at the information, since the information has become the focus of enterprise competition, we can completely regard information as a product of the enterprise, and the production of department-level products is the site of "Information workshop", The site (System) for overall management of enterprise-level information is called an information factory.



Figure 1 CIF Architecture

The CIF architecture is represented in the diagram

Therefore, a basic infrastructure such as CIF must be able to absorb and support enterprise information from different sources. Not only does it need to invest heavily in unstructured data from file management systems, internal and external Web sites, operating systems, groupware, and e-mail, but it also invests in traditional structured data (from data warehouses, data marts, operational data Stores (ODS)). The creation, use, Exchange, and deletion of all information in an enterprise should be achievable in this infrastructure.

The Enterprise information factory consists of some existing systems based on processing, these systems are connected by information, and for the normal operation of the CIF, each component on the production line must share a consistent and solid infrastructure that has specific requirements: flexibility to adapt to change, low redundancy, scalability and scalability. EDW (Enterprise Data Warehouse) – Located in the center of the Enterprise Information Factory, contains the following data

Particle size of

History of

The whole.

Of all

The data created in Edw provides information for data marts, data mining, and so on, in short, EDW is the basis for the operation of data warehouses and decision support systems. The data in the EDW is granular and without redundancy, based on the consensus that two or more people have formed through ongoing discussions about the results of different analyses on CIF basis. Edw can be stored in primary storage (high performance) and level two memory (poor performance).

The following is a brief description of the main conceptual elements of CIF

Operating system, data acquisition, data warehousing, primary storage management, level two storage, data marts, data submission (a set of view processing processes that enable end users and their information systems groups to establish and manage data warehouses, which include the following processes: filtering, formatting, and submitting data from data warehouses to data marts), ODS, Metadata management, explore the database (is a decision support system architecture, the purpose of the exploratory and temporary processing to provide a data port, it uses data compression technology to achieve a short period of time to access the entire database). The data mining warehouse, an environment in which analysts test their assumptions and inferences in the exploration database, and specialized data mining tools include intelligent agents to perform these tasks. Analytical applications (pre-designed, installable decision support applications that often need to be customized to meet enterprise-specific requirements. For example, risk analysis, CRM analysis, etc.). External Data (External) is any external data (such as demographics, credits, competitors, and financial information that is located outside of the formal data of the enterprise application, typically, the external data is provided by the enterprise to a specific information merchant for purchase)