ITERATIONS: The Data Warehouse Development Methodology

Though I've only alluded to it so far, the time has come to give the comp.databases.* faithful a preview: Prism Solutions' second major release of its data warehouse development methodology will be formally announced at DB Expo in San Francisco next week. The methodology, now named ITERATIONS(tm), envelops the practical field experience of our consulting group and close consulting partners. I personally have been proud to co-lead this effort with J.D. Welch, while collecting and assimilating the knowledge and workplans of Prism project leaders, data warehouse architects, and data modelers around the world. If you will be in attendance at DB Expo (I cannot), J.D., Nick Caffarra (Director of Consulting), or one of our marketing team will be available to discuss ITERATIONS with you. The following is the text from a one-page draft overview we've been distributing to interested customers and partners: ** Data Warehouse Specific Methodology ** Data warehouse projects pose a very unique set of analysis, design, technology and management challenges which are unlike traditional development projects. Successful data warehouse implementations require a methodology that reflects these differences to efficiently guide project managers. More than a set of technologies, successful data warehouse implementations are the result of an effective process and architecture. ** ITERATIONS: The Data Warehouse Development Methodology ** Prism Solutions' data warehouse consulting expertise is not limited to the skills and knowledge of its professionals. The ITERATIONS Data Warehouse Development Methodology bundles years of experience related to the specifics and intricacies of delivering successful data warehouse solutions. This flexible, iterative project roadmap adapts to various teaming arrangements and differing types of data warehouses across all industries. Based on Bill Inmon's original data warehouse development approach, this enhanced methodology covers all facets of the data warehouse development project from management, to modeling, to technical preparation. ** Why Another Methodology ** First, ITERATIONS is designed to supplement and graft with existing systems development life cycle (SDLC) development methodologies, not replace them. Using an SDLC methodology alone however can yield drastic results when applied to a data warehouse project. These methodologies are intended for developing transactional systems that run the business, not informational systems that report and analyze the business. Second, ITERATIONS reflects the ongoing parallel steps and iterative cycles needed to deliver successful data warehouse solutions in reasonable time-frames. Most important, ITERATIONS details activities specific to the data warehouse development process which are absent from other methodologies. ** Methodology Evolution ** Prism's data warehouse methodology has its roots in Bill Inmon's original decision support systems (DSS) development process of the 1980's. Formalized in 1992 as Prism's initial data warehouse methodology, it has continued to be refined and expanded while Prism has delivered nearly 200 data warehouse solutions. Today, encompassing the workplans, techniques and guidelines of projects that span industries, Prism is introducing its second major release of the methodology. ** Methodology Framework ** Prism Solution's ITERATIONS methodology is organized for flexibility and usability. The work breakdown structure is comprised of Modules, Activities and Tasks. Modules represent high-level work items that should be addressed to some degree in any data warehouse project. Each Module is defined by a set of deliverable-oriented Activities with recommended Tasks. Modules are organized into five process Tracks: Project, User, Data, Technical, and Meta Data, representing groupings of parallel work items. As in common methodologies, Modules are also organized into sequential sets of Phases. ** Work Item Definitions ** Each data warehouse specific Module and Activity is defined in detail with: - Summary - Detail Description - Inputs - Deliverables - Responsibility - Predecessors / Successors - Techniques - Effort Estimating Guidelines - Quality Assurance Factors - Success Criteria - Meta Data Impact ** Additional Components ** The ITERATIONS methodology is rounded out by a set of Data Warehouse Process Guides, Role Definitions & Matrix, Deliverable Definitions & Matrix, an on-line Generic Data Warehouse Workplan, and formal methodology training. * * * Please feel free to publicly post any questions you may have about ITERATIONS' content. Contact me directly (dlaney@prisminc.com) if interested in its availability. Cheers, Doug Laney, Consulting Manager Prism Solutions dlaney@prisminc.com 847-292-7100 x129 www.prismsolutions.com