Enterprise Application Service Model

Abstract

Following the recent changes in the global business environment, many organizations are reevaluating their approach to delivering enterprise applications and are looking for more effective ways to control IT costs. There is growing evidence of reluctance to fund large-scale implementation projects, and of tighter budgets forcing more careful cost-benefit analysis to justify IT investments. It is becoming increasingly clear that the traditional model for delivering enterprise applications that involves the implementation of licensed software such as ERP (enterprise resource planning) applications within end-user organizations is not suited to the fast-evolving business world of the 21st century. Almost invariably, situations in which organizations own and maintain their entire IT infrastructure lead to very high costs of ownership, and consequently high levels of IT spending, which can detract from the core business in which the organization is engaged. This has led to a situation in which some businesses doubt the benefits of IT (Carr, 2003), and some observers even contend that productivity improvements, once assumed to be the result of IT, are more likely to be the results of other factors such as longer working hours (Nevens, 2002). This backlash that followed the IT boom at the end of the last century has forced software vendors to seek more cost-effective models for the delivery of enterprise applications, and has led to the reemergence of the ASP (application service provider) model as an alternative to licensed software. Today, the ASP model (or software-as-a-service model) is a part of a more general trend toward utility computing, where the service provider delivers highly scalable application services to a large population of end-user organizations in a reliable and cost-effective manner, typically from a remote data center. Utility computing aims to supply application services on demand, similar to other utility services (e.g., gas or electricity), and relies on new technologies and architectures that enable the virtualization and sharing of resources across a large number of users in order to minimize costs and maximize utilization. The use of advanced service-oriented architectures (SOAs), grid computing, cluster technologies, and failure-resistant configurations enable the delivery of highly scalable application services in a reliable manner to a large population of users. These technological advances distinguish utility computing from the earlier ASP and outsourcing models, and will ultimately result in significant reduction in the costs of enterprise software solutions and wide adoption of the software-as-a-service model. Major IT vendors including IBM, Microsoft, Sun, Oracle, and HP are promoting utility computing, albeit under different names (e.g., on-demand computing, etc.), and are investing vast resources into the construction of data centers and related facilities (Abbas, 2003). Others, such as Salesforce.com, have been successful with providing hosted services for CRM (customer-relationship management) and other related types of applications, validating the ASP model and further confirming the trend toward utility computing. As the enterprise application software market matures, major ERP vendors are changing their revenue model to decrease their reliance on new software licenses toward income generated from software-license upgrades and product support (Karpecki, 2004; Levy, 2004). This change combined with the fact that most organizations spend as much as 80% of software-related costs on software maintenance and related activities (Haber, 2004) creates a situation in which licensed software is de facto rented. It is precisely this high level of ongoing costs that motivate many organizations toward alternatives such as outsourcing and the ASP model. In this article we first examine the business drivers for the ASP model and contrast the software-as-a-service model with the traditional software-as-a-license approach. We then discuss future enterprise computing trends, focusing on the reemergence of the ASP model for enterprise applications and the likely impact of the wide adoption of this model on the IT landscape. In conclusion, we summarize the main arguments in this article.