SOA?s Next Frontier: Information as a Service

May 30, 2006

SOA?s Next Frontier: Information as a Service

SOA?s Next Frontier: Information as a Service
By Dr. Fern Halper, Partner and Marcia Kaufman, Partner

Delivering information as a service is the next frontier in the movement to a Service Oriented Architecture (SOA). Most of the early SOA implementations have focused on creating reusable and sharable business services.  Hurwitz & Associates sees organizations beginning to apply the same architectural approach to their data.  This evolution in the technology of information management is being driven by the business need to have secure, consistent, and trusted data sources that can be used efficiently throughout the organization. By decoupling data from its tight linkages with specific business applications, business can begin to use data as a flexible resource shareable across many applications. 

Many organizations today find that their data is trapped in a complex tangle of disparate data stores and technology systems.   Multiple departments or divisions of one company might each have their own business applications and associated data stores that have been designed to meet their unique needs. For example, the sales and inventory control departments are likely to rely on different business applications and each of those applications is usually tightly linked with its own data store. Each application may need similar data, such as customer, product, or pricing data, but the definitions of these data often vary across departments. In addition, the data from the various data stores may have different structures, different interfaces, and even different semantics.  This multiplicity of applications and data sources leads to redundancy and inconsistency in data across the enterprise. In addition, it has become very difficult to manage and control the data security of these multiple data stores, particularly as organizations open more of their internal business services and data to customers and partners.

Delivering Trusted Data as a Reusable Resource

An architectural framework that provides reusable information services is an effective approach to minimizing the problems of inconsistent and duplicate data. Data inconsistencies can choke the life out of a business, slowing down business decisions making and expanding time to market. The information as a service approach is designed to ensure that business services are able to consume and deliver the data they need in a trusted, controlled, consistent, and flexible way across the enterprise. Once the data is decoupled from the business application it can be reused and shared in different ways as needed. To make information available as a service requires three steps:
1.  Create consistent data definitions.  The creation of a metadata layer is critical to establishing consistent enterprise-wide business and technical definitions of data across applications and data sources. A metadata layer is used to coordinate, track and mange the semantics and rules of an organization?s data sources.  It is also used to track changes to the data.

2.  Develop of a set of data services to qualify the data and make it consistent and ultimately trustworthy. Businesses use various software tools to ensure that when multiple sources are integrated, the data from those sources will work well together. These tools are used to profile, ensure quality, transform, audit, and otherwise create trusted data. All of these data processes can be exposed as data services so they can be shared and used across all of a company?s various data sources.

3. Create of a data architecture based on loose coupling. Retain the data in existing sources, but loosen the linkages to the applications so the data can come together in different ways as needed.

Within the next year, we expect there to be a flood of vendors offering solutions that support the information as a service approach.  Two vendors who have been doing advanced work in this area include IBM and Informatica. Both companies are providing this kind of technology in the context of their overall data integration strategy. The overall goal of these solutions is to provide business management with data that will support clear, consistent, and reliable information for their business by better leveraging common data.

Informatica?s strategy is to strengthen their product line so they are in a position to participate fully in a customer?s entire data integration life cycle.  Last years acquisition of data quality vendor, Similarity Systems, was an important step for the company to support their goal of offering a more complete and integrated solution to customers. Informatica also makes use of partnerships such as their OEM relationship with Composite to provide federation capabilities.  Informatica?s experience with customer?s early attempts at SOA demonstrated the need for data services. The company views a data service as ?a modular, reusable, well-defined, business-relevant service that enables the access, integration and right-time delivery of enterprise data throughout the enterprise and across corporate firewalls?. Informatica?s solution for SOA provides for consistent access to the data according to four key services: access services, integration services, metadata services, and infrastructure services. The data is delivered via SQL, RSS, Xquery, JMS, or via a web service to the intended application, process, or service.  The services themselves are built using Informatica?s development tools. 

IBM’s approach to information integration is to provide a broad range of interoperable integration services that plug directly into service oriented architectures. Over the past several years, IBM has made a huge investment in developing their information integration platform through acquisitions of companies like Ascential Software and SRD as well as reaching out to a broad ecosystem of partners. IBM?s WebSphere Information Server has been designed to provide the end-to-end abstracted data access layer that will pull together all the data service components. The profiling, quality, transformation, auditing, and federation functions will all take place behind the data abstraction layer. Underlying these functions is a common metadata and parallel processing infrastructure that provides automation across the platform. The metadata is linked into a common repository which provides lineage and impact analysis.  Each product also provides connections to many data and content sources, and the ability to deliver information through a variety of mechanisms. 

The key benefits of implementing information as a service include increased business flexibility, business trust in data, and reduced costs. The integrity of the data is strengthened because when a business service receives or consumes data that is delivered as a service, the data has been effectively certified by the enterprise as trusted data. The ultimate goal of this approach is to provide a seamless way for the business user to access data that is trusted and consistent with company rules and polices. Hurwitz & Associates expects more businesses implementing a SOA strategy to incorporate information services into their approach.  

Newsletters 2006
About Fern Halper

Leave a Reply

Your email address will not be published.