Nurturing the Roots of Service-Oriented Delivery
Abstract: Much has been written about the benefits of a service-oriented approach to the design of enterprise solution architectures. Many organizations looking for a strategic transformation of their information technology landscape have adopted a service-oriented approach to obtain the benefits of a more agile and flexible business service layer. However, many that have tried will fail to realize the full benefits this architectural approach due to neglecting the importance of the metadata layer in achieving agility and reducing complexity; as well as reducing the costs of delivery and overall cost of ownership. Collaborative metadata registry and ontology solutions such as Adjoovo Spaces provide a straightforward solution to realize faster and higher returns from service-oriented IT solutions.
Roots of Services
When people think of service-orientation, the focus naturally falls to the services and processes that they wish to create, compose and automate. This straightforward approach correctly targets some of the key deliverables in the future service landscape. However, these elements reside at the very top of the visible problem space, and are supported by a much larger mass of metadata from which an organization can multiply its returns if managed effectively. Comprehensive management of the metadata within an organization will significantly reduce the complexity of the process engineering and systems integration challenges that are part of any service-oriented approach. The layer referred to in this whitepaper as the “roots of services” represents this metadata layer. Nurturing it through effective and comprehensive management will multiply the benefits that an organization will reap as they walk the road to building an agile, enterprise service framework for their business.
Figure 1 depicts the landscape of service-oriented design and the relative size of the problem posed by the metadata layer. A root cause of failure in these projects is that the underlying metadata assets are largely ignored as an organization builds new solutions and services. This mismanagement - or negligence – of the importance of what exists, during the analysis, design and implementation phases leads to architectural fragmentation and much higher levels of complexity. Put simply, increased complexity unequivocally leads to increased cost.
Increased size, complexity and fragmentation drive additional costs into the development, deployment and maintenance of service-oriented solutions and critically reduce the agility of the service and process layer. More metadata translates into a larger quantity of required code and other artifacts which in turn leads to higher development and maintenance costs. Multiple entity representations (or fragmentation) leads to higher process complexity. Fragmentation and higher complexity leads directly to less agility. Less agility means longer implementation cycles and undermines the ability of the business to respond to its changing requirements.
Hidden Costs in Your Service Landscape
Few organizations have a coherent view of their enterprise data-model. The model encompasses all the entities within the business (e.g. customers, employees, products, services, incidents, events, etc.), different applications (e.g. customer relationship management, billing, product information, …) and, of course, the services and processes that those solutions might expose. Fundamental to gaining this understanding is being able to “mine” the mass of metadata that exists within the firm. Using this information to create a register of available assets, their relationships and the assets in which they are defined provides a cornerstone for a service-oriented initiative. Without such a foundation, no group or individual can fully understand this data landscape. As a result, duplicate entity representations will be created as new pieces of the final solution are constructed. This duplication only serves to add complexity to the data-model and fragments the service architecture. The end result is an ever growing spiral of complexity and an ever-decreasing likelihood of re-use; a vicious cycle from which escape becomes increasingly more challenging.
Figure 2 illustrates a common situation where duplicate representation of a business entity, in this case data about a customer, leads to higher complexity of a business process. In the example, multiple representations of the customer data need to be integrated to allow for automation of a process. Although the process orchestration is conceptually simple – a sequence – the concrete implementation has now been complicated by the required transformation and the data validation and error handling associated with it.
Although a very simple case, the cost of delivering this business process has been increased due to the duplication and fragmentation in the underlying data representation. An increase in service or process complexity adds real overhead to the service delivery lifecycle. Metadata fragmentation leads directly to reduced agility as incompatibilities in data models require special attention, thereby diverting resources. Coupled with the increased delivery costs, the increased maintenance overhead further erodes the return-on-investment an organization is able to realize. When multiplied and compounded at an enterprise-scale, the impacts can translate into thousands of wasted employee-hours and result in huge opportunity costs as solution implementation-cycles grow.
Increased complexity and cost are just some of the symptoms of poor metadata management. Ultimately, the problem can become more serious and result in poor data quality, through errors introduced by multiple transformations and not detected by weak validations. In turn, these directly risk the accuracy of business processes and services themselves and, in extreme cases, can affect the perceived operational integrity of the business and damage external perception.
Putting aside the risks to data quality, at the very least the fragmentation of the metadata layer significantly reduces the agility of the service-layer and inevitably delays the anticipated benefits of an organization's investment in a service-oriented approach.
A Better Approach
There are multiple solutions already in place that assist an organization in maintaining a “map” of the operational service layer and managing the process of taking a service through the testing phase into production use. However these tools ignore the issue of coherent metadata management. The Adjoovo Spaces registry solution is different. Its ability to analyze a wide range of development assets, both collaboratively and in real-time, allows it to create a much more complete dictionary of not only the service definitions, but the processes and, critically, the data and message types used within those higher-order artifacts. By maintaining a registry of the relationships between all of these entities, organizations can obtain a holistic view of their metadata landscape.
An Holistic View
This holistic view aids an organization's service-oriented developments in many ways. The business analyst or architect, at the outset of a new project, can quickly survey the metadata landscape to collate details of the key business entities that will be needed and the forms that represent them within the current systems, services, processes, etc. Critically, it is this degree of visibility that allows an organization to re-use its existing assets rather than duplicate and fragment. Re-use has a number of significant benefits, both immediate to the project and longer term to the organization, as discussed below.
Constructing this holistic view is aided by the ability of the Adjoovo Spaces solution to extract metadata from a wide range of development artifacts. This non-destructive approach to metadata extraction and annotation allows Adjoovo Spaces to be integrated into any existing development methodology and tool-set. Essentially, Adjoovo Spaces “mines” the metadata layer within an organization, collating and providing a platform to develop a rich and accessible enterprise data dictionary that will improve the understanding of your organizations metadata landscape.
Improving Collaboration
Visibility and awareness are keys to collaboration and Adjoovo Spaces does not stop at “mining” the metadata but also provides a framework that encourages awareness and promotes collaboration. It applies leading-edge, community-based approaches to the registry's web user-interface allowing community members to tag, document, annotate, share and search the metadata entirely in real-time. Managed by Adjoovo Spaces, all of this elaboration and collaboration is performed in a non-destructive way so that the original assets are preserved.
Higher-levels of Re-use
Re-use, leading to more consistent data representation and service definitions has other important benefits; it enables an organization to benefit from the promise of agility that service-orientation promises, but many firms fail to realize. Representing data about business entities in a standardized way allows services to be recombined more easily allowing an organization to produce new composite applications that respond to their changing business requirements with increased agility. However, before an asset can be re-used the organization has to firstly know that it exists, where and how it is currently used and its relevance to the challenge at hand. Adjoovo Spaces provides a rich set of tools for discovering existing artifacts from simple or custom keyword searching to more sophisticated semantic awareness.
Metadata Lifecycle
Adjoovo Spaces provides features for tracking the use of metadata throughout your project lifecycle. At the inception of a project, required artifacts are identified by searching the registry which can then be collated onto a “workbench” and shared between all actors involved in the elaboration of the project. As the project progresses through the planning, development, testing and production phases, each entity can be decorated with additional information such as design documents, test plans or even pointers to source code and other artifact dependencies. Each resulting artifact is tracked throughout the project lifecycle to show where it is being used, and how it may have changed over time. This resource becomes a valuable reference for future operations and maintenance activities as well as a rich base of material for future design and development projects. The enterprise data mining and annotative features in Adjoovo Spaces, when combined with its automatic and user-defined dependency tracking controls, guard against the risks associated with compromising data quality and integrity in the complex and evolving nature throughout the metadata lifecycle.
Adjoovo Spaces
Through years of experience in building service-oriented platforms and solutions, the professional services team at Adjoovo understand the complexities involved in SOA, WOA and other forms of service-orientation. They know how to adapt the correct technology, processes through their experience to optimize your service-oriented initiatives and maximize the return-on-investment. Adjoovo Spaces is a product solution realization of that collective experience and can be integrated into any technical landscape and used with virtually any delivery methodology.
The core of the product is a flexible registry and management framework which assists an organization to map out their existing service-oriented assets and elaborate their understanding of them. This provides the organization with a holistic view of those assets thereby empowering anyone with access much greater visibility into the results of their past investments. For the organization in question, the immediate benefit is to de-risk their initiative by ensuring that knowledge of these assets is cataloged and managed and not left solely in the hands of key information workers or external consultancy firms.
Acting as a corporate meta data dictionary the registry can be used in different phases of the solutions delivery process, for example:
- Business analysts can use it to understand which services or processes are available to support their transformation projects.
- Architects can use the catalog to search for existing or archived services that may resemble new project initiatives, looking for opportunities to re-use and extend existing artifacts.
- Developers can use the registry to document key aspects of services thereby enriching the registry with the collective knowledge of the organization's technical workforce.
- Operation managers can leverage the knowledge to identify critical services, identify key technical resources involved in its construction and to document service level agreements that have made with the supported business units.
These are just some of the uses that having a managed lifecycle registry like Adjoovo Spaces can bring. Overall, the benefit of such a collaborative framework is increased efficiency of the delivery process, reduced risk to the organization, and a much higher return-on-investment from your service-oriented initiatives.
Take Adjoovo Spaces for a test-drive and enable your organization to experience the benefits.