Application Architecture Strategy: Integration or Insulation?
In the 1970’s our focus as Information Technology professionals was on how to construct individual application systems. We focused on such things as structured analysis, walkthrough strategies, testing approaches, and project management techniques. We also discovered that getting end-users involved in the projects in a big way was a genuine improvement over existing practices at the time.
In the late 70’s and into the 80’s, having “conquered” the how to problem, we broadened our attention from individual applications to the suite of applications used to automate the business. Integration became the watch word of the day as our efforts focused on integration of applications across the business. This became easier to do with the advent of new, more powerful DBMS technologies and was managed using emerging IRM-based techniques. In many cases we succeeded in what we set out to do: we built highly integrated application suites to support the business.
Just as we thought that we might actually be getting on top of our the problem of IT management, our world has started to unravel in the 90’s. Client-server technologies, rapid prototyping, object-oriented techniques, rising end-user expectations with the advent of LAN’s and PC’s, rapid business change, and the coming of age of the packaged application software idea have not only challenged our “how to” solutions, but have put strain on our concepts around integration. As business requirements change, the highly integrated solutions that we have put in place are becoming barriers to business change.
In this presentation we examine strategies for the architectural design of applications that take advantage of the new trends in information technology to position for more flexible business systems which can keep pace with the accelerating pace of business change.
Some of the core ideas covered include strategies for “insulating” applications from one another so that components of the business systems can be replaced or radically modified to keep pace with the changing business requirements without affecting other components in the system. Data replication strategies combined with new transformation hub technologies take these ideas from the design drawing board and make them achievable today.
The bad news is that this represents a very fundamental change in our overall architectural design strategies. The good news is that our much of our IRM thinking and many of our data-centric techniques provide the keys to making these new strategies implementable and manageable.
Lee Stem, a founding partner with Chartwell I.R.M. Inc., has many years of experience in the development, teaching, and practice of Information Resource Management (IRM) principles and methods. He has assisted many organizations in both the public and private sectors with the development of Information Architectures and formulation of Strategic Information Technology Plans. In addition, he has worked closely with these organizations to help them develop the underlying methodologies, practices, and tools to effectively implement these plans. A primary focus in his practice has been the coupling of the plans for Information Technology with the Strategic Business plans of his clients.
Mr. Stem has extended the model-based approaches born out of the architecture practice to facilitate Business Process Re engineering projects with several of his clients. He has lead these clients through the process of fundamentally re designing the way in which they carry out their business. These highly successful projects have combined Quantitative Business Analysis techniques with expert facilitation services and have resulted in the identification and pursuit of substantial changes to the business process with significant bottom line impact.
In an information management career spanning more than twenty years, Mr. Stem’s entry into the consulting field was as a Database/Data Communications consultant for a major computing services provider. He has worked with a variety of organizations in both technical planning and project management roles. In 1980, he joined the insurance industry where he progressed from Systems Development Manager to Director, Administration and Systems. He resigned his position to join Chartwell IRM Inc. as a founding partner in 1984.