The quest to provide IT (information technology) systems infrastructures that can support the required level of end-user service availability for an organization is ongoing and likely to be never ending. That said, real progress has been made as engineers and practitioners have increased their objectives from a focus on reliable systems, to multisystem configurations, to multisite deployments, so that today it is possible to provide service that is continuously available (i.e., available 99.999 percent of the time) to end users.

This paper introduces responsive systems: systems that are real-time, event-based, or time-dependent. There are a number of trends that are accelerating the adoption of responsive systems: timeliness requirements for business information systems are becoming more prevalent, embedded systems are increasingly integrated into soft real-time command-and-control systems, improved message-oriented middleware is facilitating growth in event-processing applications, and advances in service-oriented and component-based techniques are lowering the costs of developing and deploying responsive applications. The use of responsive systems is illustrated here in two application areas: the defense industry and online gaming. The papers in this special issue of the IBM Systems Journal are then introduced. The paper concludes with a discussion of the key remaining challenges in this area and ideas for further work.

Service systems produce all services of significance and scope, yet the concept of a service system is not well articulated in the service literature. This paper presents three interrelated frameworks as a first attempt to define the fundamentals of service systems. These frameworks identify basic building blocks and organize important attributes and change processes that apply across all service systems. Although relevant regardless of whether a service system uses information technology, the frameworks are also potentially useful in visualizing the realities of moving toward automated service architectures. This paper uses two examples, one largely manual and one highly automated, to illustrate the potential usefulness of the three frameworks, which can be applied together to describe, analyze, and study how service systems are created, how they operate, and how they evolve through a combination of planned and unplanned change.

Since the onset of the computer age, customers have been searching for ways to increase systems availability, maximizing the return on their investment and keeping their systems running continuously as much as possible. Traditionally, customers have used a combination of hardware and software architecture and processes in an attempt to address their continuous availability needs. In this paper, we introduce a new approach to provide high availability which leverages technology for both proactive and reactive support, while engaging a trusted support partner to automate services. We show how automated problem reporting reduces problem resolution time, often by sending the problem to a specialist rather than sending a generalist to the problem. We describe how inventory reporting enables proactive support by applying intellectual capital and knowledge to a specific customer configuration. We also describe how the IBM Electronic Service Agent™ can be used as a foundation for providing service automation and enabling a digital relationship that expands the capabilities of the trusted support partner.

The current approach to the design, maintenance, and governance of service-oriented architecture (SOA) solutions has focused primarily on flow-driven assembly and orchestration of reusable service components. The practical application of this approach in creating industry solutions has been limited, because flow-driven assembly and orchestration models are too rigid and static to accommodate complex, real-world business processes. Furthermore, the approach assumes a rich, easily configured library of reusable service components when in fact the development, maintenance, and governance of these libraries is difficult. An alternative approach pioneered by the IBM Research Division, model-driven business transformation (MDBT), uses a model-driven software synthesis technology to automatically generate production-quality business service components from high-level business process models. In this paper, we present the business entity life cycle analysis (BELA) technique for MDBT-based SOA solution realization and its integration into service-oriented modeling and architecture (SOMA), the end-to-end method from IBM for SOA application and solution development. BELA shifts the process-modeling paradigm from one that is centered on activities to one that is centered on entities. BELA teams process subject-matter experts with IT and data architects to identify and specify business entities and decompose business processes. Supporting synthesis tools then automatically generate the interacting business entity service components and their associated data stores and service interface definitions. We use a large-scale project as an example demonstrating the benefits of this innovation, which include an estimated 40 percent project cost reduction and an estimated 20 percent reduction in cycle time when compared with conventional SOA approaches.

In this paper, we describe an approach to business processes and services which views work practices as recurrent patterns of communication called genres. Although defining work practices in this way is unorthodox, it provides two major advantages. First, the communication resources employed by the parties engaging in a service transaction can be clearly described, understood, and communicated. Business processes and services can be differentiated on the basis of the structural and functional arrangement of their constituent genres. This provides a view of a business process or service that is technology-independent. Second, using this approach means that work practices are defined contextually—an important consideration when trying to understand how business processes and services will influence organizations. Because genres are represented using directed graphs, prototypes can be developed to assist during the analysis of existing services and the design of new ones. Structural and functional change of genres can be used to reveal how a specific service is evolving within an organization. This enables us to determine if business demands have changed, something that is difficult to achieve using conventional service engineering approaches.

Designing and implementing a business resilience (or disaster recovery) plan is a complex procedure for customers, and the impact of implementing an incorrect or incomplete plan can be significant. For some customers, being able to recover their data center functionality in a short period of time may be of the utmost importance; for others, recovering in a short period of time may be worthless if the data with which their database is restored is hours or days old. Also of importance is the impact to business-critical applications when copies of data are being made. This paper presents the IBM TotalStorage™ Productivity Center for Replication (TPC-R), a tool designed to help customers implement cost-effective data replication solutions for continuous availability and disaster recovery. We give an overview of TPC-R, describe recent enhancements to TPC-R that are available on all supported platforms (as well as those that are unique to the z/OS™ platform) and discuss the ways in which customers can exploit TPC-R to implement business resilience solutions, with a focus on the various trade-offs customers must consider when choosing between different storage replication technologies.

This special issue of the IBM Systems Journal focuses on responsive systems, a term in which we include real-time and event-based systems. A realtime system is a computing system subject to constraints on the interval from the occurrence of an external event until the system response. A real-time constraint can be hard, in which case failure to complete the required computation before the deadline amounts to a system failure, or it can be soft, in which case delayed responses are tolerated but viewed as lowering the quality of the service provided by the system. An event-based system is a computing system whose components operate in response to events. The events processed by a component can be either external to the system or they can be generated by another component.

Leveraging redundant resources is a common means of addressing availability requirements, but it often implies redundant costs as well. At the same time, virtualization technologies promise cost reduction through resource consolidation. Virtualization and high-availability (HA) technologies can be combined to optimize availability while minimizing costs, but merging them properly introduces new challenges. This paper looks at how virtualization technologies and techniques can augment and amplify traditional HA approaches while avoiding potential pitfalls. Special attention is paid to applying HA configurations (such as active/active and active/passive) to virtualized environments, stretching virtual environments across physical machine boundaries, resource-sharing approaches, field experiences, and avoiding potential hazards.