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.
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.
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.
Generic Data Architecture Principles
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.
DATAXMART has formulated generic data architecture
principles that would become the guiding principles
for designers and developers in a development
Reusable Code Components
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
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.
Information Delivery Architecture
DATAXMART during projects also works towards customer
specific templates. These on going forward can
be leveraged for substantially reducing coding
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.
Data Quality Framework
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.
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.