We use cookies and other similar technologies (Cookies) to enhance your experience and to provide you with relevant content and ads. By using our website, you are agreeing to the use of Cookies. You can change your settings at any time. Cookie Policy.

Business IT Buzz Blog

Kaitlin Brunsden

Data Modeling for MDM: Talking with Embarcadero

Vote 0 Votes

What follows is my podcast with Jason Tiret, Director of Modeling and Design Solutions at Embarcadero. We discuss how the popularity of MDM has prompted a new series of questions around data modeling. Jason will offer his insight on the risks of not having a data model before launching an MDM initiative.

Listen to my 4:40 podcast below:

Download file

KB: Can you please provide us the brief overview of your company?

JT: Yes. Embarcadero Technologies specializes in heterogeneous database and application development tools. And really, we target application developers, database developers, DBAs, and data architects, and really try to maximize their productive with the systems that they're managing.

KB: Today's topic is Data Modeling for MDM. The popularity of MDM has prompted a new series of questions around data modeling. So Jason, to what do you attribute all the recent interest in data modeling for MDM?

JT: Well, there are a lot of moving parts with master data management. You're integrating a lot of data from a lot of different sources. And anytime you're doing that, you need to have your models in place so that you basically have the blueprints or the roadmap of what the systems look like, what the data looks like, and you're going to be a lot more effective, save a lot of time when implementing or beginning on a master data management project.

KB: What are the risks of not having a data model before launching an MDM initiative?

JT: Well, I think it's mainly comes down to cost and time. You're going to waste a lot of cycles if you don't have the proper documentation. If people don't understand exactly how the sources of data that have the data spread out all over the place are integrated into the master data hub, then there's going to be a lot of wasted time finding the people and the knowledge of the system to be able to integrate the data between those two things. So really, you're going to be wasting a lot of money and a lot of time. And rather than relying on tribal knowledge and going to individuals that might have information about the sources, it makes sense to have models in place so that everybody's on the same page.

KB: How are data models different for MDM and is the value any different?

JT: They're not necessarily different. But really again, you're dealing with a lot of data movement when you're talking about master data management and a lot of scrubbing of the information for master data management so you really need to have the data models of the sources. But then when you're aggregating that stuff together, have a kind of master copy of a model that takes into account all those different sources and comes up with a kind of common denominator of what they look like and that's a big reason why a lot of our customers have leveraged the universal data models that we resale. They're built by Len Silverston and a big application of those data models is for master data management because they can give you a standard of customer data, of product data, order information, HR data, whatever it might be, they give you a nice even kind of playing field for what all the different sources might have.

KB: Is it possible to use the existing data models for MDM and if not, can we tweak the ones we have to work?

JT: Absolutely, I think you're going to have to some tweaking and there's going to be some sharing that's going to occur across those different data models and its really -- the sharing might even be at the [indiscernible] level in coming up with a common set of domains that you can share across the models so that you can ensure that if you're integrating customer data, for example, that all the address information is defined the same way, or if it's not defined the same way that you come up with a common standard for that address data.

And what you're also going to be doing with existing data models is a lot of data lineage and mappings between the source target and mappings and that's very important. A lot of times I see a lot of customers managing those mappings externally from the data model and doing it in Excel, or Access, or some home grown repository where that really requires a lot more work because if the model changes then those mapping files change as well. If you can get those into the model, then you're going to be a lot more effective, save a lot of time, and maybe drive the communication from the data model and the reporting that you can do off the data model rather than having to manage an external source of that mapping information.

Keep up with what's hot in the world of business and IT integration.

Jayaprakash Kannoth

Jayaprakash Kannoth is Software Engineer at TechTarget. His areas of interest include business process management, enterprise architecture, business intelligence , cloud/infrastructure computing and technology in business.
The opinions expressed herein are my own and do not represent my employer’s views in any way.


 Subscribe in a reader

Recently Commented On

Monthly Archives