data warehouse design

Data Warehouse Design

Managing the design, development, implementation, and operation of even a single corporate data warehouse can be a difficult and time consuming task.

Problem Definition
	The definition and recording of the problem to be solved is one of the most often overlooked step of any development effort. A business problem needs to be solved, so avoid just jump right in a solve it. For Data Warehousing design, it can be a wrong approach. What specific Problems will it solve? Write down the number of problems that this data warehouse will need to solve. The problems should be clearly stated, be very specific, and have quantative criterias for success. All these problems and issues need to eb published and reviwed by the dev team. What are the available resources: time, budget and Dev resources? Defining the project resources upfront will help to complete project successfuly. We need to be realistic about available resources to avoid missing deadlines, budget overruns and low quality. Going for a data warehouse when you only have resources for a database, will lead to failure. Do we need to upgrade the existing system, perform conversion from a legacy system, or have a need to develop a new one from the scratch? Developing a data warehouse from scratch is much easy then re-engineering of existing one. Ugrading existing system or converting from legacy system will be more difficult. We need to make sure that fully understand what system we will have after the migration and how we need to integrate it with other applications. If the budget is low, then consider developing a "small" data warehouse - a data mart. With bigger budget we may develop a new data warehouse. We know that hardware, memory, and disk space are cheap. It's the software and personnel time that are expensive.
Requirements Analysis
	Performing a complete business and system requirements analysis is critical to the success. Without a clear goal and objective it's difficult to reach success. Some of the approaches stated below: Clearly state the business goal of the project and problems we need to resolve. Need to identify the data sources and incoming data format. Review the user activities of the users who will be using the future system Ask the future users to explain what they need from the future system. Define success and acceptance criteria. Create priorities of all requirements.  Create detailed project schedule including all important milestones. Be prepared to update it weekly. Create Acceptance Requirements and set ex[ectation to the same level among team members.
Design & Prototyping
	Rapid Prototyping  There are 4 important steps that can be considered: Gatjer a small team of database developers, hardware specialists, QA Engineers, tech editors, business subject experts and a Project Manager. Allocate a small group of business users who will provide the feedback necessary to drive the prototyping cycle. Generate a user manuals and design interfaces before start development. Once the system prototypes are successful, then  we can begin the development process.
Development, Unit Testing and Documentation
	Once you started the requirements analysis and have working prototypes, it's time to begin the development. Coordinating development schedule, resolving multiple project issues, and hardware installation, system development, documentation documents, managing project reviews, and testing can take 100% of your time. Hardware incompatibilities, data format incompatibilities, software bugs, late deliveries, etc. are very common during a data warehouse project execution.
	Data Warehouse Solutions · e-Business Intelligence · Re-Engineering of DW applications · CRM Data Warehouse · Data Warehousing Implementation · Data delivery applications · Data integration Case Studies · Business Management System · Data Warehouse WAP System · Cell-phone Users Survey · BI Platform Development · Data Warehouse Reporting Portal Data warehouse Technologies OLAP tools:MicroStrategy, Brio, Business Objects, Cognos, Hyperion, Informix Metacube ETL tools: Informatica, Datastage, Datajunction, DataMirror Databases: Oracle, MS SQL Server 2000, Sybase, DB2, Informix, Redbrick, Teradata; Datamining Tools: SAS Miner, Intelligent Miner; End to end tools: SAP Business Warehouse suite and Oracle Data Warehousing product suite » Offshore Application Development Framework.  Real Time Database development outsourcing