DATA WAREHOUSING ARCHITECTURE

Data Warehousing Architecture



Data Warehousing Architecture

Introduction

Data warehouse is a data base, which is a repository for storing and analyzing numerical data. Typically, the main data repository is a numeric value that can be accumulated, or group, stored in structures that are different from conventional structures transactional database. One of the reasons why the structure of the database storage structure is different from the transaction database is that data mining can be very resource-intensive operation. Data warehouse can store data in aggregate form, which means that to obtain summary values ??are not required to perform ad hoc queries that are sensitive to time and resources, such as creating a report in the transactional database. This is a simplified definition. This concept is described in more detail in books on data warehousing (Beynon, 2004).

As many users of database systems are not familiar with computers, developers hide the complexity to users through several levels of abstraction to simplify the user interaction with the system. There are different levels of abstraction to simplify the user interaction with the system: The internal level: It has an internal schema, which describes the physical storage structure of the database. The internal scheme uses a physical data model and describes all the details for your storage and access roads to the database. The conceptual level: Has a conceptual framework that describes the structure of the entire database to a user community. The conceptual schema hides the details of physical storage structures and concentrates on describing entities, data types, links, user operations and restrictions. At this level we can use a data model or a high level of implementation.

At the conceptual level the database as a collection of logical records without storage descriptors. Actually conceptual files do not physically exist. The transformation of conceptual records to store physical records is performed by the system and is transparent to the user. Level vision: Highest level of abstraction, is what the end user can visualize the finished system, describes only part of the database to the authorized user to view. The system can provide many insights to the same database (Connolly, 2002).

Discussion

Over the past 15 years, companies have spent billions of dollars on display and data storage. Everyone has long obvious that, first of all, we must clearly understand the source systems, start with one or more functional areas and business processes, but gradually expand the project to enterprise scale, and provide end users with access to data, tools and applications that meet their requirements. However, there is one question that many people still find it difficult to answer. Which architecture should I use?

Error in this way can bring serious damages. BI-architecture is defined by business requirements, and for its successful use should be guided by long-term goals, and consider what changes might be needed in the future. Solutions within the individual units must meet the current reporting needs, but also potentially expand to enterprise scale (Date, 2003).

Five Architectures

To date, proposed a set of architectures, we describe ...