Making the Case:
OMG's Model Driven Architecture
By Jon Siegel
October 15, 2002 —
(Page 1 of 6)
Information technology serves the enterprise best when it focuses on business first, technology second, but this is hard to do when your most powerful tools focus exclusively on technology. Business modeling serves the enterprise, but needs to couple directly to IT in order to deliver the goods. Object Management Group Inc.'s Model Driven Architecture (MDA) makes the connection, unifying business modeling with technology (from legacy systems to the latest and greatest middleware platforms) into an industry-standard architecture that leading enterprises are using today to build and run IT systems that keep them ahead of the competition. In this article we'll describe the MDA, starting with its foundation on world-class business modeling, and show how this extends into the technological realm to support the portability and interoperability on which today's enterprise depends.
But first, why focus on modeling? Because modeling is the only way to ensure that enterprise IT systems deliver the functionality that a business requires, comprehensive and stable, yet able to evolve in a controlled manner as business needs change over time. Models built in the Unified Modeling Language (UML) represent exactly what a business application-even a complex, multiplatform integrated application-can do, and record it with a clarity and stability that far exceeds that of the applications themselves, which change with the technological fashion. An enterprise with a repository of models has a confidence level in its IT that cannot come from any other source. Based on technology-independent representations of their business functionality and behavior, its applications last for decades and maximize IT return on investment (ROI).
If modeling is the answer, why doesn't every enterprise center its IT on a group of architects and designers today? Pre-MDA, models and code have been developed separately by different groups of people. Programmers regarded the models as guidelines or rough plans rather than firm requirements. Disadvantages abound: Compared with development without modeling, in this environment it costs twice as much and takes twice as long to produce a final application that resembles and benefits little from its model, regardless of how good it may have been.
Share this link: http://sdt.bz/26807
Most Read
Latest News
Resources
SAP unveils SAP HANA platform innovations for Big Data and spatial processing
Features include smart data access and expanded cloud deployment options
|
|
|
Alteryx raises $12 million to put Big Data analytics in the hands of all business analysts
Quest founder's firm, Toba Capital, selects Alteryx as its first analytics investment
|
|
|
Google I/O kicks off
Developers get new APIs and tools, and the Go language hits version 1.1
|
|
|
Jelastic launches new version of its Java and PHP hosting platform
Jelastic today announced the launch of a new version of its ultra-scalable cloud hosting platform
|
CollabNet fuses CloudForge, TeamForge
New pricing structure and integration gives developers an enterprise-grade choice for dist...
|
|
|
Eclipse release train for Kepler arrives June 26
New version of Eclipse includes Stardust for business process management, and Orion 3.0 fo...
|
|
|
Google I/O kicks off
Developers get new APIs and tools, and the Go language hits version 1.1
|
|
|
Enterprises going mobile get first ALM platform
Solstice Mobile releases AppLauncher for native app development and deployment on multiple...
|
IDC MarketScape: Worldwide Cloud Testing and ASQ SaaS
Demand for solutions to test applications on the cloud and for the cloud is rising signifi...
|
|
|
Get to Know the Database Decision Factors
What should you look for when choosing a relational database system? This informative arti...
|
|
|
Exploring the Database Forest
Today’s database technology landscape is more dynamic and varied than ever before. What’s...
|
|
|
Data Management Resource Guide
Today’s data is generated by more than just applications. Data is generated by trillions o...
|