Vous êtes ici

Data Warehousing & Data Modeling

Data Warehousing & Data Modeling

A well-designed and professionally built data warehouse is a cornerstone component of any best-in-class Performance Management architecture. It contains the single version of the truth for all relevant management information, historic, current or future, all of which should be of impeccable quality.

Building such a data warehouse is not an easy feat. Testimonies to this are the numerous data warehouse projects that have failed in the past, both for small and large organizations and deployments. Past project successes and failures have learned us a lot in terms of best practices in data warehousing.

The challenges for building a good data warehouse are many. The experienced consultants at element61 have extensive expertise in designing and implementing them; a successful data warehouse requires a heterogeneous project team with complementary skills, both functional and technical. element61 offers the full range of services that are required to design, build and maintain a best-in-class data warehouse with maximum Return-on-Investment for the business organization. Our skill set includes :

  • Data modeling. We understand the art of translating business requirements and business processes into a conceptual data model according to the well-established principles of dimensional modeling as defined by Ralph Kimball and co. This resulting data model represents the business process that we are targeting and is an important tool of validating the design with the business users. Key in this process is that the data modeling is done in a sufficiently generic fashion, keeping the data model open enough and making it relatively straightforward to implement changes in the future as the underlying business process evolves.
     
  • Data model implementation. It is crucial that the conversion of a logical/conceptual data model to the physical data model, implemented in the database, is done correctly. element61 has the knowledge and skills to do translation keeping into account the specific characteristics of each of the leading RDBMS platforms such as Oracle, Microsoft SQL Server, IBM DB2, Teradata, ... This involves the use of combined features such as aggregate tables, materialized/indexed views, partitioning, advanced indexes, etcetera.
     
  • Database and data warehouse tuning. High-volume inserts and updates, as well as flexible and fast query response times are a crucial factor to the success of a Business Intelligence or Performance Management environment. In order to achieve this, we apply a mix of technology-independent data modeling techniques, combined with tweaking the relevant RDBMS settings. This results in excellent data warehouse performance for all types of data interactions.

element61 can design and build your data warehouse completely from scratch, regardless of the technology used. Alternatively, we have a methodology which will enable us to quickly perform an audit of your existing data warehouse and provide you with hands-on recommendations on how to improve it, if needed.

In any case, we advise you not to underestimate the importance of good data warehouse design. The investments made in designing an open and flexible data model will easily be offset by the extra cost involved in cumbersome development and maintenance of second-rate data models, not allowing you to implement changes in the business as they occur.

Contact us for more information or with your specific request.