DATAXMART utilizes a few fundamental building blocks to ensure a sound foundation for scalable and reliable data warehouse architecture.
These critical building blocks are captured in the figure below and are detailed in the following sections:
  •  Business Driven Requirement Gathering Framework

  • DATAXMART requirements collection framework gracefully integrates role-based functional requirements and non-functional aspects of the requirements like performance, security, frequency, latency, interface, history and the like.

  •  Corporate Data Model Framework

  • Evolving a corporate data model framework helps a data warehouse to gracefully scale when the business environment changes and the requirements from the warehouse are revisited. A huge challenge in data warehousing is conformance of dimensions across the subject areas and this is a core area addressed in the DATAXMART Data Model framework.

  •  Robust Metadata Framework

  • Metadata is "data about data," a catalog of information about primary data that defines access to the warehouse. It is the key to providing users and developers a road map to the information in the warehouse and is an often-neglected area while building the warehouse. DATAXMART places significant emphasis on this and as a part of the methodology ensures that sufficient information is captured in a structured manner as a part of the development process.

  •  Generic Modeling Constructs

  • At the core of DATAXMARTS approach lies the "design patterns for data warehousing". A pattern can be described as an abstraction from a concrete form, which keeps recurring in specific non- arbitrary contexts. It is a named "nugget" of instructive insight, conveying the essence of a proven solution to a recurring problem in a given context amidst competing concerns.
    DATAXMART has developed this repository of successfully recurring "best practice" that has proven itself in the "trenches". This is also a literary format for capturing the wisdom and experience of expert designers and communicating it to novices. The generic design patterns of DATAXMART leverages on template models that are used universally as best-practice models and also on the distilled knowledge generated in earlier projects executed by DATAXMART.
    Examples of some of these generic constructs are Party-role-transaction construct, Contract construct and Recursive relationship construct. These are used by DATAXMART to provide flexibility to the solutions and to ensure that these solutions are scalable when the business environment changes.

  •  Generic Data Architecture Principles

  • DATAXMART has formulated generic data architecture principles that would become the guiding principles for designers and developers in a development project.
    Examples of this include guidance in terms of when to use procedural language vis-à-vis ETL, when to use canned reports vs. OLAP delivery vs. dashboards and when to create redundant structures like materialized views and when not to use the above.

  •  Reusable Code Components

  • DATAXMART consciously seeks to leverage templates of code for PL-SQL and ETL mappings from its component repository that has been developed over a long period of time. These reusable code components reduce timeframes for delivery and also improve the quality of the developed code.
    DATAXMART during projects also works towards customer specific templates. These on going forward can be leveraged for substantially reducing coding effort.

  •   Information Delivery Architecture

  • A robust information delivery mechanism that provides role-based access to content in the integrated data warehouse is a foundational requirement in any data warehouse implementation.
    Alternative modes of information delivery like static Web-based "bread and butter" reports for operational roles, dashboard/scorecard based reporting for senior managers, exploratory OLAP reporting for analysts, proactive rule-based alerts in cases where exception based alerts are all required, need to work in tandem in a warehouse.
    DATAXMART uses a structured approach to identify specific requirements and define a robust architecture that caters to all these requirements. Standard reusable reporting components and format templates also get defined and this eases the effort for churning out new reports.

  •   Data Quality Framework

  • DATAXMART recommends at least 4-6 core metrics to quantitatively assess the quality of measures and dimensions data in every engagement. DATAXMART would dip into the standard library that is maintained for known data quality issues.

    All the above frameworks are continuously evolved by the DATAXMART BI Labs. People in the labs are responsible for coming up with reusable components, checklists, coding standards, proof of concepts in emerging areas and best-practices repository.

    The knowledge that is accumulated is disseminated to people in projects. People in projects can also avail the help of labs resources when they have an acute technical problem to solve.
    Existence of this focused technical BI lab resources, which can be leveraged across multiple projects, is a key differentiator for DATAXMART.

    Copyright © 2006 DATA XMART CORPORATION . All Rights Reserved