Imagine acquiring, storing, analyzing, enhancing, maintaining, archiving and distributing overlapping, inconsistent, incomplete and even conflicting information of your products, customers and suppliers across half a dozen systems within your company. Then overlay system and application integration with point-to-point connections or multiple messaging flavors for consuming applications. The result is a messy, inflexible and expensive architecture that delivers problematic information to internal and external clients. If your organization stores overlapping data across systems and has not addressed master data management at an enterprise level (regardless of a Web services or messaging integration layer), you might be closer to the described antipattern than you think. ( See Figure 1, Stage 1.) In today’s challenging economic environment, CIOs are looking for cost reductions and increased agility to meet changing business demands. IT simplification powered by information and integration enablement translates to operational efficiencies, simpler regulatory compliance, overall infrastructure reduction and labor cost savings. We support the case that addressing such simplification concurrently through a combination of MDM and service-oriented architecture provides a superior architectural framework with the highest cost savings. We will concentrate on the information and integration areas of simplification with their associated impacts to infrastructure; while opportunities for IT simplification exist in other areas such as governance and interaction, they are not covered in this article. We assume the reader has a basic understanding of MDM, SOA and their intersection in a data services layer. The SOA references in this article include both Web services and messaging which leverage the same SOA infrastructure and canonical schemas, further emphasizing the point of simplification.

We will assume that systems A and C in our example (Figure 1, Stage 1) can be retired by consolidating data and authorship to an operational transactional MDM hub, while system B provides additional functionality that needs to remain in use. For a realistic view, we presume that retirement of legacy systems is accomplished through an iterative, phased approach due to scope and impact of work. We derive an interim logical diagram (Figure1, Stage 2) by introducing or promoting an existing SOA enterprise standard data distribution layer, and retiring systems A/A2 into a newly introduced MDM hub. Systems B/C/C2 feed the MDM hub or are alternatively connected through SOA directly to consumers as an interim solution in order to avoid disruption of service. Since the information of systems C / C2 needs to be migrated to the MDM hub anyway, mapping to and populating the MDM hub in the interim state is preferred to SOA enabling the legacy systems which would be a throwaway effort. We derive an end state (Figure 1, Stage 3) by retiring systems C/C2 and extending SOA integration to include data acquisition to the MDM hub, which depicts how MDM and SOA simplify the integration and improve the consistency of information delivered to consumers. 

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access