In other words, data warehouse architecture is simply a blueprint which describes the elements and specific services of the warehouse, with details showing how the components will interrelate together and how the system will grow over time. Moreover, the architecture provides the conceptual framework of the warehouse in which these components are mutually arranged specifically to suit the organizations requirements and to function in an optimum way. In order for these functions and services to be provided, a warehouse uses basic computing platform which generally makes use of the available technology.
The platform includes the various software and hardware products installed as well as the type of users and the policies that govern it. However, these elements can be categorized as either operational infrastructure or physical infrastructure. Operational infrastructure deals with on how to keep the data warehouse running. These elements would include the people, the trainings required, the policies that each govern a function, and management software that would help maintain the efficiency and management of a data warehouse.
Fundamentally, the physical infrastructure of a warehouse, as Ponniah noted, consists of the basic hardware components, the operating system with its utility software, the network, and the network software (p. 147). Coupled with a set of tools needed to provide such functions and services of individual architectural components. These components are pre-selected that may go through a number of critical evaluations in order to meet the necessary requirements to support the entire data warehouse.
Moreover, Ponniah suggests that the infrastructure has to be modular as possible. That is, when newer versions are cost-effectively available, components could easily be unplugged and plugged in the upgrade. The data warehouse computing environment consists primarily of the hardware and the operating systems which provides jobs such as data extraction, transformation, integration, and transportation. Selection of these components are passed to certain criteria, such as, scalability, technical support, security, reliability, availability, and memory protection.
Additionally, these infrastructural components would make up the front-end and the back-end systems of the entire data warehouse. However, managing these databases would need a robust and reliable DBMS that would match the selected hardware and software components. The DBMS shoul also have the capability of delivering a balanced data output and portability to access across the different platforms. Software tools are also important in every architectural component of a data warehouse.
Third-party software tools can provide the necessary needs for developing a data warehouse computing environment, such as, data modeling, GUI design software, query tools that would generate reports, data warehouse administration and others. Generally, these tools cover all the major functions of a data warehouse.
Ponniah, Paulraj (2001). Data Warehousing Fundamentals: A Comprehensive Guide for IT Professionals. New York: John Wiley & Sons, Inc. Orr, Ken (2000). Data Warehouse Technology. Retrieved from the Web March 9, 2007. http://www. kenorrinst. com/dwpaper. html