Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 1
--> Overview and Summary Crises are extreme events. They cause significant disruption and put lives and property at risk. Some crises arise from natural disasters such as earthquakes, hurricanes, fires, and floods. Man-made crises can be accidental, such as oil spills or the release of toxic substances, or they may be intentional, such as bombings by terrorists. Crises require an immediate response and a coordinated application of resources, facilities, and efforts beyond those regularly available to handle routine problems. Crisis management was the primary application area examined in the Workshop Series on High Performance Computing and Communications conducted by the Computer Science and Telecommunications Board of the National Research Council (see Box S.1). Crisis management was selected not only because of its critical importance to public safety and well-being, but also because building good tools that are useful in meeting the extreme demands of crisis management requires significant advances across a combination of many different, broadly applicable computing and communications technologies. The challenges confronting crisis managers are extreme in several dimensions. Crises require an extraordinary quantity of resources, such as search and rescue teams, medical assistance, food, and shelter. The demands are highly diverse—implying a need for cooperation among many different actors—and largely unpredictable in terms of location, time, and specific resources needed. Moreover, the urgency associated with crises has many implications, such as the need to rapidly identify, collect, and integrate crucial information about the developing situation; to have access to tools and resources that are not cumbersome or difficult to use, particularly in stressful conditions; and to have the capability to make projections and
OCR for page 2
--> BOX S.1 Overview and Context of the Workshop Series Consultation with people who use or want to use computing and communications to accomplish their objectives provides a sometimes sobering perspective on technology design and implementation. The three Computer Science and Telecommunications Board (CSTB) workshops on high-performance computing and communications were designed to foster discussion among application specialists-including technology experts (developers or systems managers) and professionals in crisis management, digital libraries, electronic commerce, manufacturing, and health care-and researchers to explore how computing and communications technologies are used in these areas, the problems or shortcomings associated with current technologies, and potential improvements that might both enhance the technology base in these national-scale applications and advance the state of the art in computing and communications. Researchers and users discussed not only traditional high-performance concerns, such as speed and scale of computation and networking, but also capabilities in information management, collaborative work, decision making, and many other areas. Such capabilities are enabled by advances in the underlying computer and network systems and at the same time make them more useful, thus hastening the evolution of a collection of computers and communications links into an information infrastructure. The workshop series fits with the intent of several federal programs to foster greater interaction among researchers, developers, and users of leading-edge computing and communications. The framework for many of these activities has been the High Performance Computing and Communications Initiative (HPCCI; see Appendix B for a brief discussion), which has stimulated such interactions, beginning with scientific investigation of ''Grand Challenges" and continuing toward study of broader "National Challenge" applications. A CSTB review concluded that the HPCCI has demonstrated the value for computing and communications research of interaction between developers and users of technologies (CSTB, 1995). Related activities include a February 1994 forum involving several hundred researchers and others, "R&D for the NII: Technical Challenges," that yielded a discussion of current research topics in communications and computing infrastructure technologies (Vernon et al., 1994). The Committee on Information and Communications (CIC) of the National Science and Technology Council developed a plan for coordinating research and development (R&D) across multiple federal agencies, identifying strategic R&D focus areas that relate to agency missions and other user needs (CIC, 1995). More narrowly focused efforts have illuminated research opportunities in specific application areas, such as health care and digital libraries.1 These recent examinations of research needs, however, have drawn mainly from the research community. By comparison, the CSTB workshop series emphasized crisis management as an application domain and featured the substantial participation of end users, including nontechnologists.2 Traditionally, crisis management has not been a focus of academic computing and communications researchers, other than in the context of military system development. But at the CSTB workshops, crisis management inspired fresh discussion of a full range of computing and communications research issues and provided a real-world perspective for calibrating research needs related to other national-scale applications—some of which have been examined more extensively through various federal programs and private-sector activities-against one that is particularly demanding in terms of urgency and unpredictability of needed resources. Crisis management was also appropriate for framing questions relating to federal support for research in an application area that is primarily a public-sector responsibility. 1 For examples, see Davis et al. (1995) and Lynch and Garcia-Molina (1995). 2 This mix contributed to the evolution, subsequent to Workshop I, of an NSF workshop focused on health care (Davis et al., 1995). A participant identified individuals in Workshop I to invite to the NSF event, as he reported in Workshop III. initiate actions in the face of an inevitable degree of uncertainty and incompleteness of information. Workshop discussions covered a spectrum from research through development to deployment and use of technology. The mix of professionals fostered consideration of how the conditions in which computing and communications are used can affect the perceived value of technologies and the demand for improvement—nontechnological conditions, too, shape perceptions about the kinds of features that would be helpful. Resource constraints of local and state crisis management agencies, for example, limit the amount of training available to users of technological tools and require users to trade-off performance and other features of new technologies against the life-cycle cost of equipment. Out of these discussions came ideas about where truly high performance technology may be helpful in different application domains, where advances in performance at the leading edge would yield benefits in more mainstream
OCR for page 3
--> systems, and how the interaction of applications in different areas through the evolving information infrastructure, on a national scale (e.g., the use of telemedicine and digital libraries in crisis response), influences the development and use of computing and communications. This changing context highlights the need for fundamental research to understand cross-cutting problems arising in national-scale applications (see Box S.2) that, on a smaller scale, may appear merely to be questions of implementing known technologies. Workshop participants agreed that the formulation of research questions by researchers will benefit from an explicit recognition that the technologies arising from today's research will be deployed to meet real needs. Chapter 1 discusses unmet demands for computing and communications technologies in crisis management and four other national-scale application areas—digital libraries, electronic commerce, manufacturing, and health care. Computing and communications technologies are increasingly central to manag-
OCR for page 4
--> BOX S.2 Characteristics of National-scale Applications National-scale applications such as those discussed in this report—crisis management, digital libraries, electronic commerce, manufacturing, and health care—use computing and communications on a nationwide (even global) scale.1 "National scale" therefore implies the potential for large volumes of computation and communications, a large number and diversity of individuals and organizations, and the associated complexity. National-scale efforts such as crisis management are distributed across multiple locations, are often linked by networks, and make use of a variety of computing and communications resources. The people involved vary in expertise, ranging from scientists and engineers to citizens who may lack specialized technological knowledge. Because they operate in such a broad and diverse environment, computing and communications systems for these application areas must be able to survive and adapt to variety and rapid change in the needs of individuals and organizations for technologies. The national-scale applications examined in the Computer Science and Telecommunication Board's three workshops have several elements in common: Scale. National-scale applications raise qualitatively new challenges for computing and communications technologies because of the geographic distribution, extent, and diversity of requirements for processing, storage, and communication of information, as well as the number of interconnected end points-users, computers, and information sources and repositories. Demand for dependability. As people come to rely increasingly on the computing and communications systems that serve national-scale applications, these systems begin to become part of the infrastructure society counts on, as the telephone system did early in the century. Consideration must be given to systems' survivability, security, fault tolerance, and graceful degradation (as opposed to catastrophic failure), among other issues. Systems must also respond to the rapid, continuing evolution of underlying technologies in a competitive marketplace. Architectural diversity. Distributed ownership of systems among many organizations and individuals makes monolithic, rigidly defined architectures largely impractical.2 Common interests among parties in one application domain such as health care or banking may result in agreement on specific architectural elements, but these interests evolve, and so generality and flexibility are required. This requirement for generality and flexibility implies the need for common interconnection standards, as well as support for people to identify and integrate the resources available to them across multiple system architectures. Heterogeneous interfaces and standards. National-scale applications are distinguished by an enormous degree of heterogeneity and decentralization in the interfaces and models of interaction among the systems that support them. Centralized control or widespread agreement on a set of protocols and functional interfaces is difficult to achieve.3 In designing systems to support broad national activities with many autonomous players, the determination of where, what, and how much to hold in common among system elements must be an ongoing process open to diverse—and competing—commercial implementations.4 1 Although it is not the case that each instance of using these applications is necessarily distributed across the entire nation, one of their distinguishing characteristics is that a given use potentially may draw upon resources anywhere in the nation—in some cases, the world. 2 Architectures are the underlying models of systems and how they relate to each other. Interfaces and standards generally embody a particular architecture; however, a standard may be used in more than one architecture, and an architecture may have more than one implementation. 3 This is true even in a relatively centralized institutional context. For example, heterogeneity is quite evident in military systems, particularly when more than one service branch or national force is involved. In addition, because the Department of Defense and other users of large-scale systems are relying increasingly on available commercial technologies, the choice and definition of standards are becoming increasingly significant for them. 4 The need for openness and evolution raises questions about how to achieve formal and informal standards and conventions on a national scale. The continuing need for standards to achieve interconnection and integration suggests the broad value of research that can clarify choices among alternative technologies while those technologies are being developed, can increase compatibility among technologies, or can generate new technologies that diminish the problems associated with heterogeneity. ing activities in all of these areas. However, current state-of-the-art technology is not always adequate to meet existing and emerging demands. Society's dependence on information technology is not absolute; certainly, fire fighters can continue to put out fires without computerized maps, and doctors can write clinical reports with pen and paper. However, continued improvements in the quality, efficiency, accessibility, and dependability of nationally important industries and services are realizable through advances in information technology and their integration into the work practices of organizations and individuals. Consequently, whether expressed as needs of society or as opportunities for researchers, unmet demands for improved capabilities in areas of broad national significance suggest many fruitful problems for research in and development of
OCR for page 5
--> high-performance and other computing and communications technologies. These research opportunities are discussed in detail in Chapter 2. Not all of them are new; however, from the perspective of crisis management, problems familiar from other nationally important application domains take on an added dimension because of requirements for systems and applications that are flexible across extremes of scale, diversity, and rapid change. Chapter 3 presents the steering committee's findings based on inputs from the workshop series and a sampling of additional, related sources. Box 3.2 presents selected examples of compelling, applications-motivated computer science and engineering research topics identified in discussions between crisis management experts and technologists at the workshops. These include
OCR for page 6
--> communications resources such as rapidly deployable, self-configuring wireless networks for coordinating response teams; "judgment support" tools to assist crisis managers in making decisions in the absence of complete, reliable information; simulations of phenomena such as hurricanes and fires that could deliver useful results to crisis managers rapidly; virtual "anchor desks" that place network-based resources such as simulations and information systems at the disposal of crisis managers; and other specific tools and technologies that appear promising for crisis management. Finding 1 emphasizes the importance of experimental testbeds for crisis management-related research and development. Testbeds that provide a realistic application setting, such as simulation- and field-based training exercises, can serve as demanding implementation environments for new technologies and sources of feedback to identify and refine research objectives. Application users, such as federal, state, and local civilian crisis management personnel, should participate in testbed activities. Their input is essential to assess the fit among systems, tools, and users' needs and to ensure that technology is focused on usable, practical solutions. To secure the full benefits of application-specific computing and communications technologies, there must also be recognition of the increasingly interconnected nature of national-scale applications. In application areas such as crisis management, digital libraries, electronic commerce, manufacturing, and health care, the widespread interconnection of computing and information resources and the people who use them over networks has made it feasible, and increasingly common, for resources to be called on in unforeseen ways. Crisis management, in particular, illustrates the value of being able to integrate highly diverse resources whose usefulness in an unusual situation could not have been anticipated in advance. Unfortunately, technologies developed to meet a specific application requirement often do not function well in unforeseen circumstances because of complex, difficult problems of interoperation, performance, and scaling up. Therefore, the findings resulting from this workshop series also address research, development, and deployment efforts that can lead to both consistent architectural approaches that function on a national scale and general-purpose tools and services that facilitate rapid, ad hoc integration of systems and resources. Finding 2 highlights the importance of investigating the features of existing national-scale infrastructures for specific applications to identify what features do and do not work. Findings 3 through 11 identify technological leverage points for computing and communications research investments, based on needs of national-scale applications. These findings emphasize research challenges in four areas: (1) support of human activities (e.g., improved ease of use of technologies for individuals and groups), (2) system composability and interoperability, (3) adapting to uncertainty and change, and (4) performance of distributed systems. Outcomes of testbed and architecture-study activities
OCR for page 7
--> suggested in Findings 1 and 2 should inform future reexamination of these research areas, which represent the best understanding of a range of technology and application experts in 1995-1996. The research questions discussed in this workshop report can and should motivate the scientific and engineering research communities in the future. They have the potential to increase the ability of individuals and organizations to make the most of important applications, to present intellectually stimulating challenges for researchers, and to promote significant advances in the state of technology. REFERENCES Committee on Information and Communications (CIC). 1995. America in the Age of Information: Strategic Implementation Plan. National Science and Technology Council, Washington, D.C., March 10. Computer Science and Telecommunications Board (CSTB), National Research Council. 1995. Evolving the High Performance Computing and Communications Initiative to Support the Nation's Information Infrastructure. National Academy Press, Washington, D.C. Davis, Larry S., Joel Saltz, and Jerry Feldman. 1995. "NSF Workshop on High Performance Computing and Communications and Health Care." Report of a workshop, December 8-10, 1994, Washington, D.C. Available on line at http://www.umiacs.umd.edu:80/users/lsd/papers/nsfwork.html. Lynch, Clifford, and Hector Garcia-Molina. 1995. "Interoperability, Scaling, and the Digital Laboratories Research Agenda." Report on the Information Infrastructure Technology and Applications (IITA) Digital Libraries Workshop, Reston, Va., May 18-19. Available on line at http://www-diglib.stanford.edu/diglib/pub/reports/iita-dlw. Vernon, Mary K., Edward D. Lazowska, and Stewart D. Personick (eds.). 1994. R&D for the NII: Technical Challenges. Report of a symposium, February 28 through March 1, Gaithersburg, Md. Interuniversity Communications Council (EDUCOM), Washington, D.C.
Representative terms from entire chapter: