The Rise of Architecture
We marvel at the unbelievable sophistication and beauty of buildings, civil infrastructure and military technology created more than 2000 years ago. There was a precision in the architecture and the technology of the Romans unrivaled for almost 1500 years after their demise.
The key to their success was the birth of a new science: architecture. And the lessons learned more than 2000 years ago are just beginning to find their way into the world of IT.
Many modern companies today are faced with chaotic systems, cobbled together over many years, and often times the result of mergers and acquisitions where it was simpler to just continue using the old systems than to change them.
The unfortunate result is disconnected processes, incorrect or late data, and a general lack of benefit from information technology that, in other industries, is providing a true competitive advantage. These problems are sometimes compounded by an IT department which is, well, less than receptive to change.
Over the past five or so years, we have seen the emergence of a new generation of technology leadership, one focused on a cohesive architecture which is both flexible, scalable and cost effective. Many larger IT organizations have created the position of chief architect, and recently one of the better known IT trade publications changed its name to IT ARCHITECT , reflecting the growth in importance of architecture in the field of IT.
Groundbreaking work at Carnegie Melon University in the field of information architecture and ways to measure the sophistication of both architecture and developmentthe capability maturity modelhas significantly reduced the overall cost of IT development, operations and maintenance.
Unfortunately, much of this groundbreaking work has yet to find its way into our businesses, and the result is we are not fully realizing the benefits of technology.
The first step in creating true value from you information technology is an absolutely objective analysis of your current overall situation. I would recommend this take the form of a tripartite analysis:
Application Architecture. Understanding your current data flows and how they interact with the processes is critical to knowing whether you are achieving a competitive advantage or even a basic ROI from your systems.
The science of application modeling has changed dramatically over the past few years, and there are new modeling tools such as use case models and data flow diagrams which go a long way towards understanding the relevance and timing of your information.
I teach a graduate course in system design and we spend a lot of time discussing the concept of logical modeling (based on the business, not the technology) and how, once you capture the business requirements in the logical model, you can then translate them into a physical model by extending the model using these modern modeling techniques.
The key is these modern tools model the business not the technology.
Clearly there is a technology component to the application architecture, but it should flow logically from the business model. The logical model, once validated by the business, is then translated into a physical model based on the specifics of the infrastructure and also on the technological competence of the organization.
The beauty of this two-step processlogical model and physical modelis they can be decoupled. The logical model embodies the business processes, which is the core institutional knowledge of the organization. The creation of a physical model will increasingly be outsourced based on the latest technology.
Infrastructure Architecture. A lot of change has take place here in the past five years also, new networking technologies are the essential underpinning of the modern far-flung corporation. A movement towards monolithic (a single application doing everything) and single instance (one database, and one system for the entire company) is taking place and the infrastructure necessary to support this is much different from those of the older, distributed systems.
The convergence of data, voice and video on the same network has also changed the environment. Costs for data networking have been dropping as have hardware costs and a movement to open systems like Linux has also dropped the overall operating costs of many datacenters.
I would venture that most companies do not have an architectural plan of their infrastructure that is up to date. I see many companies who dont even know the topology of their network, nor a detailed list of desktops and their current revisions.
Knowing your physical architecture intimately is the first key to improving it.
Organizational Architecture. This may be the most important analysis most companies can perform, and one of the most painful.
Looking at both the governance model and the organization model and comparing that against the level of talent available in the organization is really key to achieving a high value from the information technology expenditures in any modern organization.
Frankly, the leadership of the organization and the talent acquisition that flows from that leadership is really the key, and there are few organizations who are equipped to really step back and analyze their organizational structure and the level of their talent.
This is a painful analysis and one that I believe is difficult for any organization to do on their own because of the natural myopia of organizational culture.
There is certainly a role for the human capital management function of an organization to assist here, but frankly, its difficult for most HR organizations to adequately assess the technical capabilities of an IT organization. They can use very effective tools to judge temperament, leadership and assist in the growth and succession planning within IT but, overall, I think the "human architecture" requires a joint effort of outsiders and the HR organization.
The key here is to find a trusted advisor to impartially analyze these three dimensions, and to be able to objectively digest and act on findings, conclusions and recommendations that come out of the analysis.
Using new modern modeling tools and creating objective models against these three dimensions is the first step in creating a true competitive advantage from IT and is essential to creating a vision of how to achieve increasing ROI from future technologies.
Daniel Gingras has been CIO of five major companies and is a partner at Tatum Partners, a nationwide professional services organization of senior-level technology and financial executives who take on leadership roles for client companies. He has more than 30 years of IT experience and teaches system design and technology strategy at Boston University and CIO leadership at Carnegie Mellon University. He can be reached at firstname.lastname@example.org.