By Michael Shaw,
Managing Director of Data Engineering, Mike is a solution architect, developer, engagement manager, and thought leader in distributed technologies for big data analytics and cloud computing.

“Dance like no one’s watching. Email like it will be printed on the front page of the New York Times.” Cheeky quote, but let’s take it one step further… create business data like it’s going to be seen by the SEC, industry regulators, your bank, the company that wants to acquire you, hackers, business partners, supply chain vendors and, if no one else, your own IT department. 

Almost any operational activity you can imagine creates data. Someone needs to be in charge of it. Someone needs to define its structure, how it’s going to be stored and secured, where and for how long, and on and on. This is the role of your data architect. His or her job is to create a data architecture for your business.

 

What is Data Architecture & How Can it Protect Your Organization?

Like great moves on the dance floor, data architecture doesn’t just happen. 

The term “data architecture” is a wide-ranging topic, used to describe the concept of designing and building data standards and models and then setting a vision for eventual interactions between them. 

Consider the following example. You’re in the financial services business where you manage investments for millions of customers. You get approached by an insurance company that wants to sell life insurance to your customers. It could be a profitable partnership. To execute on the plan, though, you’ll need to create profiles of prospects for the insurance product and share them with the insurance company. 

Creating a database of life insurance prospects could mean taking customer names and addresses and appending demographic information like age, sex and marital status. You don’t want to include any private information, like social security numbers or names of minor children in the database, and so forth. What starts out as a simple list-building exercise can end up in a scandal as someone in your company accidentally posts your entire customer list, including private information, on a public cloud server. Hacking ensues, and the lawsuits follow. 

How can you prevent such an episode, which is just one of many that can result from deficient handling of data? 

In the insurance example, data architecture would set out standard principles for the customer data schema as well as for how it might be integrated with data from an outside entity, like an insurance company. The data architecture would define standards for securing the data and establish processes for ensuring that confidential data not be shared in an insecure way. This goes far beyond basic data modeling. Data architecture encompasses data models, but it also aligns with data use by applications. It addresses issues that arise with data in storage, data in motion and data flow. 

Data architecture is also responsible for establishing the parameters of data storage at the systemic and physical levels including creation of a data “blueprint” that enables the construction of a sound data design across the company. 

 

Defining Your Data Architecture Framework

The market offers a number of frameworks for defining and implementing data architecture. On a related front, there is a professional designation now for the data architect. Thus, you have a person, or people, working with a framework and strategy to realize a meaningful data architecture for your business. 

The Zachman Framework is one example. The data architecture is part of the broader Zachman enterprise architecture framework. Consisting of five layers, it comprises the following elements:

Layer

View

Data (What) 

1

Scope/Contextual

Matters and architectural standards that are important to the business, e.g. the financial services company partnering with an insurance company, as well as related business rules and policies

2

Business Model/Conceptual

A semantic model of data (Conceptual or Enterprise Data Model), e.g. organizing the data elements

3

System Model/Logical

Enterprise or Logical Data Model, e.g. columns and tables

4

Technology Model/Physical

Physical Data Model, e.g. partitions and CPUs

5

Detailed Representations

Actual databases, e.g. Microsoft SQL Server, etc.

Zachman Framework Diagram

What our experience has shown is that attention to the layers of the Zachman Framework, or the like, is essential for success with data in a business context. It may seem obvious, for example, that you need an actual database to store data. In real life, though, with highly complex IT scenarios, the selection of the database may actually become a critical success factor for an entire project. 

If you are not sure if you have a data architecture, you probably don’t. If you are wondering if you need one, you probably do. This is where Clarity Insights can help. We have the power to ingest your data from across your organization, store it in a data lake, call upon it whenever needed, and integrate powerful applications to create a seamless and comprehensive platform. If you’re interested in learning more about how to plan or build your modern data architecture, let’s talk

You may also like:

Data Architecture

No More Data Fabrics. It’s Time for the Data Trampoline.

Another day, another data management metaphor... Gartner, for example, cited the “data fabric” as one of their “Top 10 D...

Big Data Data Architecture Data Warehouse Modernization

Top 3 Modern Data Architecture Mistakes

Transforming legacy data warehouse architecture to support real-time data processing is a massive undertaking, and it's ...

Customer Analytics Data Architecture Data Warehouse Modernization

What does a successful modern data architecture look like?

The promise of advanced analytics capabilities is incredibly tantalizing, but actually reaching that level of sophistica...