The implementation of a Business Data Warehouse implies fully knowing the business business processes, therefore the incremental methodology was born with the purpose of defining indicators that provide relevant information for decision-making, reducing costs and speeding up the implementation of the project implementation of a Data Warehouse.
Introduction:
Consolidating business information today is a critical issue that all companies have to go through, since currently there is an important need for information to be consolidated in such a way that it allows us to make strategic business decisions that differentiate us from the competition in increasingly competitive markets.
The main idea is that we handle and analyze large volumes of data but with information that satisfies business needs as accurately as possible and in the shortest time.
Therefore, the Data Warehouse appears as a process to consolidate and manage data from various sources in order to meet those critical business needs.
But today, its implementation is a problem, since there are various providers and with them various methodologies depending on the strategic market to be reached.
Choosing an adequate implementation methodology implies a thorough understanding of business business processes in order to be able to design indicators and metrics that provide relevant information, as well as integrate and combine data from different business systems for strategic decision-making. cost reduction and short-term implementation to enjoy the benefits that this will bring to the company in the shortest possible time.
The implementation methodology that we will describe here summarizes all these benefits and more, since it allows us to build a Data Warehouse in an Incremental way.
Said methodology is conceived to manage and scale the dimensional complexity of an implementation project, where it will be possible to gradually enjoy the benefits that the implementation brings with it with short-term results.
Thus reducing risks and ensuring that the size of the project remains manageable in each phase of implementation, even if the eventual goal is to reach an enterprise Data Warehouse.
This methodology, however, is a global framework that considers all the aspects of creating a Data Warehouse, the idea is to maintain the conceptual framework, the methodology and the requirements in such a way that the methodology fits for all the implementation needs no matter which one. be the type of business.
This methodology is divided into 4 phases which are:
1. Definition
The definition phase is the first step of the implementation, and consists of the statement of the project objectives, to confirm, document, and prioritize the level of business requirements for an adequate implementation of the solution.
The requirements are established and prioritized in the requirements definition meetings, from which high definition business models are obtained that will allow us to capture the scope of the solution in the shortest possible time.
It is important in this phase to define the initial plan and the members that will make up the project team.
2. Modeling
The success of the modeling phase consists of verifying that the development team has understood the business processes and the requirements established in the definition phase. Therefore, regular meetings with end-user leaders and the development of a functional architectural prototype, as a means of communication between them, will be necessary to confirm that the developers have understood the business requirements.
The architectural prototype will also be used, to validate the architecture model and thus reduce development risk. Here it will be necessary to create a development environment to support these activities.
Additionally, the solution testing and administration team must be defined for a review of the architectural model. As far as possible, this team should be made up of members of the user teams who master the subject and highly expert developers.
3. Construction
The construction phase aims to obtain a solution that optimizes the business requirements. In this phase, an iterative process is used to refine the components developed for the acquisition of information and access to it, until they satisfy all the business requirements.
Once planned, the iterations must be completed by each partition and must be integrated within the management components. Finally the complete solution is tested. User leaders will provide details for the duration of these iterations.
Finally, a performance test and the creation of documentation materials to support the solution is required in this phase.
4. Production
This phase ranges from the installation of the system until it is finally working in the production environment. A final validation and loading of the initial information is required to begin managing the growth and support of the solution.
The application support is provided temporarily by the client's development team or also called post-production support team.
Periodic data backups are essential in this phase.
Keywords:
- Data Warehouse is a set of integrated data oriented to an integrated domain, that varies over time and is not transitory, which supports the business decision-making process. Iteration, as the mechanism by which we obtain the parts and the parts of the parts. Prototype, is a scale model of the real project, but not so functional that it equates to a final product, providing early feedback from users about the system.
Bibliography:
Inmon, WH, "Building the Data Warehouse", John Wiley & Sons Inc. USA, 1992.
Data Warehousing Technology, Ken Orr Institute, Jim Keogh, J2EE Reference Manual, Mc Graw Hill, Interamericana de México, SAU 2003.
