Maximizing the business value from silos: Service based transformation with service data models
Ravi Kumar Gedela, Fareign Duyu, Mark R. Cullen, Jeff Cadavez, V. Kamakshi Prasad · 2011
Legacy applications on the Mainframe platform have been, and will probably continue to be, the backbone to many large scale enterprise systems in banks and insurance companies in NA, EU and other regions across the globe. Since 1980, majority of the development, however, has been on new platforms. Predominantly, distributed systems and Web-enabled applications to support new business lines, for example, Creditor Insurance, Customer Relationship Management, and Mobile Banking, are taking the center stages. Two challenges are fundamental: how we effectively leverage the existing assets in terms of logic and data, and how we sustain the changes. The former is focusing on integration, the latter infrastructural modernization. Service Oriented Architecture (SOA) is taking hold in the industry as the preferred new way of integration and modernization. One big question remains: is SOA hype, like a fad in fashion, which will likely die out as the snow melts, or a new technological innovation alluding to financial derivatives that led to the financial market crush? Our research reveals that it is likely a hype unless a roadmap based on engineering principles is established and implemented. This paper reviews some of the failure and success factors across the industry and proposes a template to develop such a roadmap. What differentiates this paper from others is, however, a set of pressures or pain points, technical and business drivers, proven solutions and enablers (SDEP) that the authors summarized from their in-the-field and on-the-project experiences. Central to the factors is a canonical data model, which we touch on in this paper and will cover in depth in the next paper.