Executive Summary

America is entering the Age of Information. The future of U.S. leadership in advanced computing, communications, and information technology will be determined by long-term research and development conducted by government, industry, and academia. The Federal High Performance Computing and Communications (HPCC) Program continues to be a driving force for advancing these technologies and their application to challenges in science and engineering. The Program is cited as a model for multi-agency planning and coordination and helps provide the foundation for future investments in information and communications R&D that is the cornerstone of the emerging Global Information Infrastructure (GII). This report complements "America in the Age of Information,Ó the Strategic Implementation Plan prepared by the Committee on Information and Communications (CIC) of the National Science and Technology Council (NSTC), to which the HPCC Program reports.

HPCC advances enable U.S. leadership in science, mathematics, and engineering; economic growth and job creation; enhanced national security; improved environmental quality; a healthier, better educated citizenry; and effective harnessing of information technology. Information technologies are critical to the worldwide scientific community, our country's economic competitiveness, our national defense, and the quality of life of every American citizen. As the pace of advancement accelerates and competition increases, it is vital that we sustain U.S. leadership and accelerate the application of these technologies throughout our society. The Federal government plays a crucial role in this effort -- the HPCC Program, working with industry and academia, is helping to meet the mission needs of participating Federal departments and agencies and the broader needs of the citizenry.

Congress authorized the HPCC Program with bipartisan support when it passed the High Performance Computing Act of 1991 (Public Law 102-194). Created as a dynamic R&D program, it has provided the sustained focus needed for developing these technologies and has adapted to the needs and opportunities of a changing world. The Program's major accomplishments, which are described in this document, provide the foundation for R&D the country needs as we prepare to enter the 21st century.

Since its inception, HPCC Program goals (listed in Section IV) have focused on R&D in a wide range of high performance computing and communications technologies, including interdisciplinary and long- range problems for enabling solutions to an evolving set of computationally intensive Grand Challenges -- fundamental problems in science and engineering that have broad economic and scientific impact. In addition, the Program focuses on developing technologies to enable the information-intensive National Challenges -- fundamental applications that have broad and direct impact on the Nation's competitiveness and the well-being of its citizens, and advanced prototyping of a future integrated and interoperable GII capable of supporting societal needs.

Numerous reports, studies, and interactions influence the Program, including those of the participating departments and agencies, the National Research Council (NRC), the Government Accounting Office, industry, and academia. Responding to a request from Congress, in February 1995 the NRC released the report, "Evolving the High Performance Computing and Communications Initiative to Support the Nation's Information Infrastructure." That report affirms the value of the Program, its contributions to the advancement of high performance computing and communications, and the need for increased emphasis on R&D of technologies for a large-scale, integrated information infrastructure.

As Dr. John H. Gibbons, Director of the Office of Science and Technology Policy, has said, "if information is power, HPCC will be the key to power for us as citizens and for the Nation as a whole." The lessons learned in the HPCC Program from attacking the Grand Challenges and from understanding the requirements of the National Challenges will enable the development of the GII.

In FY 1996, 12 Federal departments and agencies will participate in the HPCC Program by coordinating their R&D activities and accelerating technology transfer into key computationally intensive and information-intensive applications areas. The estimated FY 1995 HPCC Program budget for the nine participating Federal organizations was $1,038 million. For FY 1996, the President requested $1,143 million for the 12 organizations participating in the HPCC Program. The 12 organizations are:

ARPA Advanced Research Projects Agency

NSF National Science Foundation

NASA National Aeronautics and Space Administration

DOE Department of Energy

NIH National Institutes of Health

NSA National Security Agency

NIST National Institute of Standards and Technology

VA Department of Veterans Affairs

ED Department of Education

NOAA National Oceanic and Atmospheric Administration

EPA Environmental Protection Agency

AHCPR Agency for Health Care Policy and Research

The National Coordination Office (NCO) for High Performance Computing and Communications coordinates the activities of the participants through the High Performance Computing, Communications, and Information Technology (HPCCIT) Subcommittee of the CIC. On March 1, 1995, the founding Director of the NCO, Donald A.B. Lindberg, who served concurrently as Director of the National Library of Medicine at NIH, completed the successful term that he began on September 1, 1992. A new full-time NCO Director, John C. Toole, was announced jointly by Dr. Gibbons and Dr. Anita K. Jones, Chair of the CIC. The Director will coordinate the HPCC Program and the emerging information and communication initiatives as directed by the CIC.

The HPCC Program is the cornerstone of the CIC Strategic Implementation Plan, part of a dynamic planning process involving government, industry, and academia. That plan identifies six Strategic Focus Areas for R&D that will benefit the Nation's diverse users of information:

Global-Scale Information Infrastructure Technologies that build advanced application building blocks and widely-accessible information services
High Performance/Scalable Systems to support "high performance" and "low end" applications in a seamless fashion
High Confidence Systems that will provide the availability, reliability, integrity, confidentiality, and privacy needed by the Nation's emerging ubiquitous information infrastructure
Virtual Environments and simulations that will continue to transform scientific experimentation and industrial practice, and will play an increasingly important role in education and training
User-Centered Interfaces and Tools to provide easier development, navigation, "mining," and general use of information resources
Human Resources and Education both to educate the next generation of industrial and academic leaders in information science and technology, and to establish a foundation for new learning technologies

While the full impact of HPCC Program accomplishments may not be known today and may not be felt until the next decade, the Program's careful and balanced investments have already led to significant achievements. Some highlights from the many accomplishments reported in this document are:

Based on research in software and networking technology, HPCC- funded communications technologies form the basis of the Internet, which continues its phenomenal growth in size, speed, and number of connected users.
A combination of gigabit-speed (billions of bits per second) networking technology and computational science have demonstrated that massive scientific calculations can be executed across parallel processing systems located more than 1,000 miles apart. These speeds represents a thousand-fold improvement over the fastest networks in existence in 1991.
Scalable parallel high performance computing systems are being developed that today represent performance that is a thousand-fold improvement over the most advanced systems available in 1991; have regained world benchmark records with speeds of more than 250 billion floating point operations per second; and promise improvements leading to meeting the goal of a trillion operations per second (teraops) performance on large scientific and engineering problems
HPCC software investments have been substantial. The National HPCC Software Exchange was established in 1994 to serve as a single software distribution point for peer-evaluated software, including new languages and tools, and benchmarking information.
HPCC organizations manage more than a dozen high performance computing facilities that house parallel, vector, and scalar systems ranging from small to large scale; hybrid systems; workstations and workstation clusters; mass storage systems; visualization systems; and heterogeneous systems connected by high speed networks. These facilities provide environments in which systems performance can be analyzed and improved and new science can emerge. One example is the virtual reality CAVE, in which users are immersed in a manipulable three-dimensional virtual environment.
Grand Challenges are fundamental problems in science and engineering that have broad economic and/or scientific impact and whose solution can be advanced by applying high performance computing techniques and resources. Use of HPCC technologies to address these problems has resulted in qualitative leaps in scientific understanding and represents a new paradigm of multidisciplinary research involving collaborative teams of computer and computational scientists and subject matter experts to solve more complex, more realistic problems in timely ways.
In 1995, the Program began demonstrating select National Challenge applications and leveraging the HPCC investment through these demonstrations. National Challenges are fundamental applications that have broad and direct impact on the Nation's competitiveness and the well-being of its citizens, and that can benefit from the application of HPCC technologies and resources. Results from R&D in the Grand Challenges and in information infrastructure technologies are being integrated into these National Challenge applications to improve our productivity, the quality of our environment, and the quality of our lives.
Technologies that directly support the missions of participating organizations are being demonstrated. For example, design change time has been significantly reduced using mediator technology, demonstrating improvements essential for streamlining Department of Defense weapons system acquisition and affordable design of large systems in industry.
The HPCC Program has instituted educational programs and new curricula that have brought computational science into classrooms at all levels and increased the diversity of students pursuing scientific disciplines.

As cited by the NRC study, it often takes 10 to 15 years for publicly-funded research to reach commercial success. This document highlights a number of high performance computing and communications research efforts, begun a decade or more ago, that are now having profound scientific and economic impact on the Nation -- examples include the Internet and R&D in global climate change and in weather forecasting. These examples point out not only the fundamental long-term contributions that federally supported research in information technology plays, but also the need to maintain fundamental research with long-term objectives in these critical areas.

The NRC report cited "the importance of retaining the HPCCI's [HPCC Initiative] momentum at just the time when its potential to support improvement in the nation's information infrastructure is most needed." The HPCC Program has provided a single national focus for high performance computing and communications technologies. This position is strongly endorsed by CIC's Strategic Implementation Plan and by industry groups.

As it goes forward, the Program will help create the very fabric of tomorrow's information-oriented society. Much remains to be done before a vigorous information economy can emerge. Systems and networks must have significantly higher throughput, bandwidth, security, and assurance to support the needs of the diverse user communities. The same infrastructure that spurs business will also carry education to both traditional learners and a new class of next- generation students. Leading-edge computational science and engineering must continue to discover new knowledge and develop new capabilities using the most advanced computing systems and communications networks. Digital libraries and collaboration techniques developed under the HPCC Program will be important enabling applications. Software must be easier to develop, execute, and adapt to all types of users. Computing systems must operate in more distributed environments. Both industry and government must be able to employ the most advanced information technology in the most affordable way possible. And, as always, well educated students will remain a critical component of technology transfer.

Our country must maintain leadership in computing and communications technologies, while applying them to our society through the Grand and the National Challenges. This in itself is not an easy task, and innovation in basic research, development, and the application of information technology will continue to yield significant new tools, techniques, and affordable applications in the decades to come. Solutions will rely on new system architectures, on advances in our understanding of the fundamentals of computational techniques and networking technologies, and on the ability of these endeavors to capture the insight and imagination of strong, well-trained minds. Only a carefully coordinated Federal effort, coupled to industry, academia, and the citizenry, can lay the long-term foundation for the Nation's information future.

NOTE: This book takes the form of a printed version and an on-line version available over the Internet. The on-line version was prepared using tools developed under the HPCC Program. It contains "hypertext links" to additional information that are identified by URLs (Uniform Resource Locators) in the printed version. These URLs change over time, and the on-line version will attempt to provide the most current links. That version can be found at the HPCC Program's "home page" whose URL is:

http://www.hpcc.gov/