Adam Deane

SOA Definition
Service orientation is a means for integrating across diverse systems. Each IT resource, whether an application, system, or trading partner, can be accessed as a service. These capabilities are available through interfaces; complexity arises when service providers differ in their operating system or communication protocols, resulting in inoperability. Service orientation uses standard protocols and conventional interfaces-usually Web services-to facilitate access to business logic and information among diverse services. Specifically, SOA allows the underlying service capabilities and interfaces to be composed into processes. Each process is itself a service, one that now offers up a new, aggregated capability. Because each new process is exposed through a standardized interface, the underlying implementation of the individual service providers is free to change without impacting how the service is consumed.

Why SOA?
Complex, distributed IT resources are a concern for businesses. Too frequently, the existing IT portfolio does not adequately meet specific business needs, is costly to manage and maintain, and is inflexible in the face of business growth and change. The solution, however, is not to rip and replace systems or applications, nor to completely renovate them, but rather to find a way to leverage existing IT investments so that overall organizational goals are effectively supported. Service orientation helps to accomplish these goals by making systems more responsive to business needs, simpler to develop, and easier to maintain and manage. Implementing a solution architecture based upon service orientation helps organizations plan ahead for change, rather than responding reactively.

Who does SOA?
Strictly speaking, SOA is done by developers and solution architects. However, stakeholders in a service-oriented solution span a range of roles, and it is critical that their interests not only be taken into account but that they actively drive the design of the SOA solution. Starting with those interests, the business analyst is concerned with bringing IT investments more in line with the business strategy. For the developer, this means that the SOA solution must map the sources of business information-systems, staff, trading partners-into a unified and comprehensive view such that the business analyst has greater insight into the costs and benefits of various investments. The chief technology officer (CTO) of the organization will work with developers to ensure that when designing a solution to meet the needs of the business analyst, the integrity of existing IT systems and applications resources are preserved, even as new capabilities are developed. And the IT manager, concerned with effectively integrating distributed systems such that management is simplified, will work with the developer to ensure that these goals are also met. Ultimately, the developers and solution architects are concerned with creating dynamic collaborative applications that meet the goals of the various stakeholders. The service orientation approach enables them to do so in a way that meets the needs of the organization as a whole.

What SOA isn’t
There are numerous misconceptions about what SOA is-that it is a product that can be purchased (it is not; it is a design philosophy that informs how the solution should be built); that the goal is to build a SOA (it is not; SOA is a means to an end); or that SOA requires a complete technological and business process overhaul (it doesn’t; SOA solutions should be incremental and built on current investments). SOA is also often equated with Web services, and the terms used interchangeably. While it is true that SOA is made easier and more pervasive through the broad adoption of Web services–based standards and protocols, the two are distinct. SOA is an approach to designing systems—in effect the architectural drawings or blueprint—that directs how IT resources will be integrated and which services will be exposed for use. In contrast, Web services is an implementation methodology that uses specific standards and language protocols to execute on a SOA solution.

Before starting a SOA
Before a developer writes a single line of code, it is critical to identify both specific business drivers of the SOA endeavor and the dependencies between the business and the underlying technologies. Neglecting the business context can result in a project in which SOA infrastructure is pursued for its own sake, or where investments are made that do not line up well with the needs and priorities of the business.

Two approaches are commonly pursued for implementing SOA: top-down and bottom-up. Both approaches have possible pitfalls that can prevent success. Many organizations that have attempted to roll out SOA infrastructure through a top-down approach have discovered that when the infrastructure is finally delivered it is out of sync with the needs of the business. Likewise, a bottom-up approach can fail as well, because it can lead to a chaotic implementation of services created without regard to organizational goals. The “middle-out” approach is a successful hybrid of the two other approaches. Business drivers and strategic vision are first employed to set clear direction and priorities. Based on these, the organization takes multiple iterative steps to build out slices of end-to-end capabilities, with each iteration delivering a new, dynamic application back to the business that is used to create business return. Microsoft has long advocated this “real-world” approach to leveraging service-oriented architectures: The approach is focused on rapid time-to-value, and it delivers business results through iterative, incremental steps that facilitate close alignment of IT resources with changing business conditions.

What is the SOA life cycle?
The core IT assets of any organization include its data, legacy systems, line-of-business applications, packaged applications, and trading partners. Each of these resources is a service provider responsible for producing numerous highly specific outputs, such as inventories and customer data.
Service orientation ties together these disparate and autonomous sources of information, bridging a wide range of operating systems, technologies, and communication protocols. The process by which it does this is an iterative one of creating (“exposing”) new services, aggregating (“composing”) these services into larger composite applications, and making the outputs available for consumption by the business user.