Person:

Maccormack, Alan

We test a method that was designed and used previously to reveal the hidden internal architectural structure of software systems. The focus of this paper is to test if it can also uncover new facts about the components and their relationships in an enterprise architecture, i.e., if the method can reveal the hidden external structure between architectural components. Our test uses data from a biopharmaceutical company. In total, we analyzed 407 components and 1,157 dependencies. Results show that the enterprise structure can be classified as a core-periphery architecture with a propagation cost of 23%, core size of 32%, and architecture flow through of 67%. We also found that business components can be classified as control elements, infrastructure components as shared, and software applications as belonging to the core. These findings suggest that the method could be effective in uncovering the hidden structure of an enterprise architecture.

The modern industrial corporation encompasses a myriad of different software applications, each of which must work in concert to deliver functionality to end-users. However, the increasingly complex and dynamic nature of competition in today’s product-markets dictates that this software portfolio be continually evolved and adapted, in order to meet new business challenges. This ability – to rapidly update, improve, remove, replace, and reimagine the software applications that underpin a firm’s competitive position – is at the heart of what has been called IT agility. Unfortunately, little work has examined the antecedents of IT agility, with respect to the choices a firm makes when designing its “Software Portfolio Architecture.” We address this gap in the literature by exploring the relationship between software portfolio architecture and IT agility at the level of the individual applications in the architecture. In particular, we draw from modular systems theory to develop a series of hypotheses about how different types of coupling impact the ability to update, remove or replace the software applications in a firm’s portfolio. We test our hypotheses using longitudinal data from a large financial services firm, comprising over 1,000 applications and over 3,000 dependencies between them. Our methods allow us to disentangle the effects of different types and levels of coupling. Our analysis reveals that applications with higher levels of coupling cost more to update, are harder to remove, and are harder to replace, than those with lower coupling. The measures of coupling that best explain differences in IT agility include all indirect dependencies between software applications (i.e., they include coupling and dependency relationships that are not easily visible to the system architect). Our results reveal the critical importance of software portfolio design decisions, in developing a portfolio of applications that can evolve and adapt over time.

We test a method for visualizing and measuring enterprise application architectures. The method was designed and previously used to reveal the hidden internal architectural structure of software applications. The focus of this paper is to test if it can also uncover new facts about the applications and their relationships in an enterprise architecture, i.e., if the method can reveal the hidden external structure between software applications. Our test uses data from a large international telecom company. In total, we analyzed 103 applications and 243 dependencies. Results show that the enterprise application structure can be classified as a core-periphery architecture with a propagation cost of 25%, core size of 34%, and architecture flow through of 64%. These findings suggest that the method could be effective in uncovering the hidden structure of an enterprise application architecture.

We propose a network-based methodology for analyzing a firm’s enterprise architecture. Our methodology uses “Design Structure Matrices” (DSMs) to capture the coupling between components in the architecture, including both business and technology-related elements. It addresses the limitations of prior work, in that it i) is based upon the actual architecture “in-use” as opposed to planned or “idealized” versions; ii) identifies discrete layers in a firm’s architecture associated with different technologies (e.g., applications, servers and databases); iii) reveals the main “flow of control” within an architecture (i.e., the set of inter-connected components); and iv) generates measures of architecture that can be used to predict performance.
We demonstrate the application of our methodology using a novel dataset developed with the division of a large pharmaceutical firm. The dataset consists of all components in the enterprise architecture, the observed dependencies between them, and estimated costs of change for software applications within this architecture. We show that measures of the architecture derived from a DSM predict the cost of change for software applications. In particular, applications that are tightly coupled to other components in the architecture cost more to change. The analysis also shows that the measure of coupling that best predicts the cost of change is one that captures all direct and indirect connections between components (i.e., it captures the potential for changes to propagate via all possible paths between components). Our work represents an important step in making the concept of enterprise architecture more operational, thereby improving a firm’s ability to understand and improve its architecture over time.