Services Computing

The term “service” has existed for thousands of years along human history. When a person or a group performs some work to benefit another, it becomes a service. Many versions of definitions exist for the term “service”. For example, James Fitzimmons¹ defines a service as follows:

“A service is a time-perishable, intangible experience performed for a customer acting in the role of co-producer.”

Chapter 1 introduces and defines some basic concepts of Services Computing; this chapter will discuss the emergence and necessity of Services Computing in the context of Electronic Business (e-Business) evolution. Referring to enterpriselevel business solutions and infrastructures over the Internet, e-Business has become a popular term with high exposure in the modern society. In retrospect, however, its original inception in the mid-nineties did not cause a sensation. Rather, as shown in Fig. 2.1, e-Business started as a common technical innovation aiming at supporting simple Internet browsing and interaction utilizing Hyper Text Transfer Protocol (HTTP). In order to enable global access, companies publish static information (e.g., business name, address, contact information, company history) on their Hypertext Markup Language (HTML) homepages on the Internet.

A Web service^[1]is a programmable module with standard interface descriptions that provide universal accessibility through standard communication protocols. The functions offered by Web services can be implemented in different programming languages on different platforms. Meanwhile, Web services can be composed to build domain-specific applications and solutions.

In Chapter 3, Universal Description, Discovery and Integration (UDDI) is introduced as typical specification for Web services registries. In general, as illustrated in Fig. 4.1, a Web service can be published in one of two major ways: to a centralized services registry or to a distributed services registry. The UDDI registry is one typical example of a centralized services registry, whereas the Web services Inspection Language (WS-Inspection or WSIL) is an example of publishing a Web service as a distributed document.

SOA stands for Service-Oriented Architecture. It is the fundamental architectural model that supports the overall paradigm of Services Computing from architecture perspective. As proved by the history of software and system development in the last fifty years, software architecture plays an essential role in software systems, by providing plausible insights, triggering the right questions, and offering general tools for thoughts. An architectural model is a representation of blueprint that contains certain building blocks common to all similar applications, along with certain variable aspects unique to each specific application. Building “conceptual constructs” is critical not only for a single software design but also for all large-scale applications and systems.

Interface-based services discovery discussed in Chapter 4 is actually a functionoriented services discovery. In the real world, business services discovery typically has to consider other features in addition to business functions. Business relationship is an important one. For example, consider that a business organization A intends to decide from two business services serving the same functionality: service b produced by business B and service c produced by business C. Assuming that business A has formed an alliance relationship with business B, business A thus will be more likely to choose service b due to this existing business relationship. For another example, suppose an enterprise E₁ needs to compose a business process including service s. Enterprises E₂ and E₃ both provide similar service s. However, there is a partnership between E₁ and E ₂ leading to a service discount, and there is no specific relationship between E ₁ and E ₃ . If price is a requirement of consideration for E ₁ , the partnership between E ₁ and E ₂ shall be counted in order to form the most appropriate business process.

As discussed in the previous chapters, the foundation of the Web services paradigm is a set of emerging standards that enable seamless integration between heterogeneous information technology processes and systems. A variety of standardization organizations and leading industrial organizations have been collaborating on Web services standards. Among them, the World Wide Web Consortium (W3C), the Organization for the Advancement of Structured Information Standards (OASIS), the Internet Engineering Task Force (IETF), United Nations Centre for Trade facilitation and Electronic Business (UN/ CEFACT), WS-Integration (WS-I) are selected standardization bodies. W3C develops interoperable technologies (i.e., specifications, guidelines, software, and tools) to lead the Web to its full potential. OASIS is a non-profit, global consortium that drives the development, convergence, and adoption of e-Business standards. IETF is a large open international community concerned with the architectural evolution and the smooth operation of the Internet. UN/CEFACT covers worldwide policy and technical development in the area of trade facilitation and e-Business. WS-I is an open industry group formed in 2002 to promote Web services interoperability across platforms, operating systems, and programming languages. Among industry vendors, IBM, BEA, and Microsoft are leading industry organizations and key contributors. With their joint efforts, currently the Web services field has developed a stack of standard protocols, categorized into a five-layer structure: transport, messaging, description/publication/discovery, Quality of Service (QoS), and service composition.

An enterprise in the twenty-first century can seldom stand alone any longer; instead, it usually needs to collaborate with its suppliers, partners, and customers on the value chain for a common goal. As the scales of enterprises grow larger and larger under the pressure of world-wide competition, the outsourcing model becomes an irrepressible trend. Instead of hiring people to do everything in house, an enterprise tends to outsource some tasks to other business entities to be more efficient, while performing the most competitive business services by itself. Meanwhile, instead of spending time to find end customers before starting a task, an enterprise also takes predefined tasks outsourced from other enterprises. Moreover, a modern business application is typically comprehensive enough to require broad specialties. While it is difficult for every single enterprise to possess a broad range of skill sets and products, it is more feasible that a set of enterprises each hold particular specialties (i.e., special services), while they together establish a virtual enterprise to provide high-quality products or services to the market sooner than what individual parties would have accomplished by their own. This kind of collaboration is different from traditional business collaboration, because it is usually project-based and service-based. As a result, not only does such a service-based value chain become more complex, but it is also usually formed dynamically instead of statically. On-demand business collaboration in a modern service-to-service value chain^[1] thus requires a more structured yet flexible collaboration adaptability.

Business companies are typically driven by underlying business processes, each referring to a set of activities that are coordinated to achieve a certain business goal. Although there are various definitions for business processes, three keywords have been widely used: tasks, flow, and business goal. A business process always implies an integration of sub-processes, or so-called tasks, each being fulfilled by individual business entities or role players. These tasks are usually organized in an activity flow that specifies a specific integration order for the tasks, either parallel or sequential, guarded by particular conditions and rules. Above all, all tasks and the activity flows serve the same business goal for the entire business process. Summarizing these key points, we define a business process as a structured and measurable set of activities that consume certain resources and are designed to produce the specified output for a particular business requirement.

When most people think of a Grid, a picture comes to mind as an interconnected system for the distribution and sharing of electricity, supported by a network of high-tension cables and power stations. Around 1995, this concept of electronic grid was applied to the field of distributed computing and parallel computing to facilitate sharing of computing power and storage resources over computers on a network. An example definition for Grid and Grid Computing is as follows:

“AGrid is a collection of distributed computing resources available over a local or wide area network that appears to an end user or application as one large virtual computing system. The vision is to create virtual dynamic organizations through secure, coordinated resource sharing among individuals, institutions, and resources. Grid computing is an approach to distributed computing that spans not only locations, but also organizations, machine architectures, and software boundaries to provide unlimited power, collaboration, and information access to everyone connected to a Grid.”

Why enterprise models are important? In general, there are two major reasons: the dynamics nature of service ecosystem that makes the modern business more sophisticated and the requirements from decision makers who ask for better internal communication.

According to Goyette and Lamar, “an enterprise involves an amalgamation of interdependent resources (including people, processes, facilities, and technologies) organized to obtain a strategic advantage in support of mission or business objectives.” One can see that the operational model of a modern enterprise has become significantly different than that of a traditional enterprise. The major cause is that enterprises have to survive unprecedented challenges, such as fluctuating market environments, world-wide competition, ever-changing customer requirements, increasingly demanded collaborations across enterprise boundaries, and disruptive technologies.

With the development of global economy, the business components in enterprises are becoming more complex. At early stage of a business entity, the boundary between business components could be quite clear and simple. However, when the enterprise grows and become large-scale, some of the business components may become overlapped with others, as shown in Fig. 15.1.

Services delivery is not a simple activity, nor is it the duty of a single role. It is the lifecycle of a service provider delivering services to a service customer as committed in a previously signed contract to satisfy the customer. The goal here is to deliver the services in an efficient way with high quality, and secure the service provider’s profit. During the services delivery lifecycle, a service provider would organize relevant resources such as people with necessary skills, using relevant software tools. It goes far more beyond project management by including setting up the IT governance before hand so that the services delivery team could follow a common principle. Common good practices covered by the governance includes asset reuse and compliant with service-oriented architecture (SOA). The lifecycle comprises of services delivery readiness phase, services delivery creation phase and services delivery operation phase.