Post No.1- Major Components of Data Architectures
Now more than ever, organization has started to see data as a
strategic asset that can be sold and exchanged. In the near future, the digital
revolution with concepts such as the Internet of Things (IoT) will push
companies to collect, and analyze vast volumes of data and therefore, architect
in a way that adds to its business value.
The enterprise data architecture should be defined and
modeled at the three levels of abstractions:
- Conceptual level to describe the high level business perspective of the data entities, how they deliver value and how they will be used. At this level data can be seen as Business information conceptual entities whether it is structured data, enterprise contents or taxonomies.
- A logical level that describes in much detailed modeling how data is represented and interrelated. At this level Schemas are defined and data are mapped to applications or taxonomies.
- A Physical or implementation level modeling that reflects the builder perspective such as Physical data stores and repositories for both structured data and enterprise wide information contents.
Without a proper data architecture, organizations won’t be
able to drive strategic change. Therefore, it is very important not to see data
architecting as a technical job. Old data models are no longer sufficient to
fulfill business demands. Experts suggest the following eight components to go
into the building of a modern data architecture:
- Engage business users in identifying the most valuable types of data
- Make data governance a first priority
- Ensure your data architecture is not developed around a specific technology.
- Develop a real-time foundation to support analysis and movement
- Build security within the foundation
- Develop a master data management strategy
- Position data as a service to enable information to be pulled from multiple sources.
- Offer self-service environments to allow business users to build their own queries.
The link below contains useful information on the topic:
Post No.2- Data Architecture Governance
Data architectures does not involve only designing data
models but it includes the rules, policies, and standards that govern how data
is used, stored, managed and integrated within an organization.
Gartner defines data governance as “the specification of
decision rights and an accountability framework to encourage desirable behavior
in the valuation, creation, storage, use, archiving and deletion of
information. It includes the processes, roles, standards and metrics that
ensure the effective and efficient use of information in enabling an
organization to achieve its goals. “
Data governance addresses many pain areas that face
organizations today. It can improve the organization ability to deploy data for
external use. It can also improve Data error remediation process and thus
enhance data quality by addressing root causes of data issues.
A data governance document in an organization typically
includes the following sections: Roles & Organizations, Data Strategy, Policies
& Standards, Compliance, Issue Management, Projects & Services, Data
Asset Valuation and Communication.
The emerging digital business that deal with vast amount of
data requires data governance to cover areas such as: Big Data, Mobile Data
Platforms, Social media, Data Demand Management and Regulatory Coordination.
Typical roles in Data governance are: Steering Committee, Data
Governance Sponsor, Data Governance Head, Data owners and Data Stewards.
Data architects may recommend enabling technologies to
support data governance and stewardship workflows such as Data profiling, Data
discovery, and Business glossary and data lineage.
For more on this interesting topic, please pay a visit to
this link:
Post No.3- Data Architecture and Informational
Architecture
I have recently read a blog whose author is among a team
that work to clean up the Wikipedia pages dealing with Enterprise Architecture.
The author was discussing how his team found out two separate pages in Wikipedia
one dedicated for Information architecture and the other for Data Architecture.
The author made a simple survey to determine whether they should keep Data
Architecture and Informational Architecture as two distinct terms in Wikipedia
or as a single term, and in this case, which one should they keep Data
Architecture or Informational Architecture.
The results of the 55 respondents was split even. Most of those
who claimed that it is one field, favored Information Architecture over Data
Architecture.
I went through different articles and blogs over the internet
that have different interesting viewpoint on the topic. Some would argue that Information
is data within context, therefore data architecture is a subset of Information
architecture. Others see Data architecture as the bigger picture whereas how
information is modeled as being part of it.
My personal viewpoint is that it depends on the context on
which the term is used. In different levels of abstraction for instance the
terms can be Data or Information. To Business Intelligence experts working to
extract data from its sources it is no doubt data, whereas it is information
for those working on Enterprise Content management system for instance. In
general, Data architecture represent the technical view and Information
architecture represent the business view.
I certainly go with the term data when I am setting architectural
principles, stewardship requirements and other governance requirements. However,
regardless of any debates on the topic, the most important thing is that Data
and Information architects should know how to build their models, methods, standards
and governance in support of the business and its drivers.
Here is the link of the blog and the survey results:
See you the in the next blog!
I like your post. The technological environment found in organizations today varies greatly. Each environment has unique characteristics that must be understood in order to effectively architect and integrate the enterprise. Most large organizations have a heterogeneous mix of legacy and current technologies from a variety of vendors.
ReplyDelete