Welcome to NASA HPCC!

You're here because you need or want an explanation and overview of the NASA HPCC Program, its mission, and how it implements and utilizes tax payer assets.

INSTRUCTIONS: You may click on the Table of Contents item (below) you're interested in and go directly to that subject. You do have the option of scrolling through the entire document which is organized according to the Table of Contents. You may return to your starting point by clicking on the ''back'' option of your browser (i.e. Mosaic or Netscape) at any time.

Please send your comments and/or questions directly to Larry Picha (lpicha@cesdis.gsfc.nasa.gov) at the Center of Excellence in Space Data and Information Sciences.

• Introduction
• The Speed of Change
• Components of the NASA HPCC Program
  • Computational Aerosciences (CAS) Project
  • Earth and Space Sciences (ESS) Project
  • Information Infrastructure Technology and Applications (IITA) component
  • Remote Exploration and Experimentation (REE) Project
• NASA HPCC Program Contributions
• Teraflops: What is it??
• Importance of NASA's Role in the National HPCC Program
• Resources: pointers to more HPCC related documentation

In recognition of the critical importance of information technologies, the United States Government created the High Performance Computing and Communications (HPCC) Program in 1991. The goal of the Program was to foster the development of high-risk, high-payoff systems and applications that will most benefit America. The NASA HPCC program is a critical component of this government-wide effort; it is dedicated to working with American businesses and universities to increase the speed of change in research areas that support NASA's aeronautics, Earth, and space missions.

By investing national resources in the NASA HPCC Program, America will be able to maintain its worldwide leadership position in aerospace, high-speed computing, communications, and other related industries. Although the High Performance Computing and Communications budget is a small percentage of the NASA budget, it has a significant impact on the Agency's mission, as well as on U.S. industry.

NASA leads the planning and coordination of the software element of the Federal High Performance Computing and Communications (HPCC) Program and is also an important participant in the National Information Infrastructure initiatives. NASA's HPCC Program will:

• Further gains in U.S. productivity and industrial competitiveness - especially in the aeronautics industry;
• Extend U.S. technology leadership in high performance computing and communications;
• Provide wide dissemination and application of HPCC technologies; and
• Facilitate the use and technologies of a National Information Infrastructure (NII) - especially within the American K-12 educational systems.

RETURN to the Table of Contents

As we stand on the threshold of the 21st century, change has become a constant in our lives. We live in a time of unprecedented social, political, and technological change and advancement. For many Americans, the rate of change has accelerated to the point where it is nearly overwhelming. It took four hundred years between the development of movable type and the creation of the first practical typewriter. Less than one hundred years later came the development of the word processor. Now, if you buy a personal computer, the computer seems to be behind the technology curve before you even carry it home from the store.

Many American business communication tools that are taken for granted today, such as FAX machines, electronic mail, pagers, and cellular phones, were unknown or generally unavailable just ten years ago. At no time in history have humans been required to process information from so many different sources at once. There can be no doubt that in the late twentieth century, the advance of technology has reached a sort of critical mass that is propelling us headlong into a future that was unimaginable a generation ago.

The rapid development of computers and communications has ''shrunk'' the world. The United States is an active participant in a worldwide economy. In this new ''global village,'' the rapid movement of information has made the technological playing field for most industrialized nations very competitive. For the first time in history, the means of production, the means of communication, and the means of distribution are all based on the same technology -- computers.

A unique interdependence now exists among advanced information technologies. Each new innovation allows existing industries to operate more efficiently, while at the same time, opens up new markets for the product itself. Individuals, corporations, industries -- even entire economies -- depend more than ever on information technologies.

America's future and the future of each citizen will be deeply affected by the speed with which information is gathered, processed, analyzed, secured, and disseminated.

NASA has a long history of developing new technologies for aerospace missions that later turn out to have far-reaching effects on society through civilian applications. For instance, satellites originally developed for space exploration and defense purposes now carry virtually all television and long-distance telephone signals to our homes.

By accelerating the convergence of computing and communications technologies, the NASA HPCC Program expects to play another unique role in shaping the future of every American.

RETURN to the Table of Contents

• Computational AeroSciences (CAS)
• Earth and Space Sciences (ESS)
• Remote Exploration and Experimentation (REE)
• Information Infrastructure Technology Applications (IITA).

The goal of the CAS project is to accelerate the development, availability and use of high-performance computing technology by the U.S. aerospace industry, and to hasten the emergence of a viable commercial market for hardware and software vendors to exploit this lead.

The goal of the ESS project is to demonstrate the potential afforded by high-performance computing technology to further our understanding and ability to predict the dynamic interaction of physical, chemical, and biological processes affecting the solar-terrestrial environment and the universe.

The goal of the REE project is to develop and demonstrate a space-qualified computing architecture that requires less than ten watts per billion operations per second.

The goal of the IITA component in the NASA HPCC Program is to accelerate the implementation of a National Information Infrastructure through NASA science, engineering and technology contributions.

Fact sheets on each of these projects are included in this brochure World Wide Web page.

RETURN to the Table of Contents

The CAS Project is focused on the specific computing requirements of the United States aerospace community and has, as its primary goal, to accelerate the availability to the United States aerospace manufacturers of high performance computing hardware and software for use in their design processes. The U.S. aerospace industry can effectively respond to increased international competition only by producing across-the-board better quality products at affordable prices. High performance computing capability is a key to the creation of a competitive advantage, by reducing product cost and design cycle times; its introduction into the design process is, however, a risk to a commercial company, that NASA can help mitigate by performing this research. The CAS project catalyzes these developments in aerospace computing, while at the same time pointing out the future way to aerospace markets for domestic computer manufacturers.

The key to the entire CAS project is the aerospace design and manufacturing process. These are the procedures that a manufacturer carries out in order to move from the idea of a new aircraft to the roll-out of a new aircraft onto the runway. Computer simulations of these aircraft vastly shorten the time necessary for this process. These computer simulations, or applications as they have come to be called, need immensely fast computers in order to deliver their results in a timely fashion to the designers. CAS supports the development of these machines by acquiring the latest experimental machinery from domestic computer manufacturers and making them available as testbeds to the nationwide CAS community. The computer manufacturers and independent software vendors help out by providing system software that forms the glue between the applications programs and the computer hardware. These are computer programs like operating systems that make the computer function.

The CAS community that carries out this work consists of teams of workers from the major aerospace companies, from the NASA aeronautics research centers and from American universities. The focus of the project is derived through extensive interactions with business managers of the major aerospace companies and by consultation with university researchers and NASA management. The project delivers applications and system software that have been found through its research to show an enhancement to the design process, and provides a laboratory by which the computer manufacturers can identify weaknesses and produce improvements in their products.

If you are interested in additional information on this project or related activities you may access the CAS Home Page on the World Wide Web. or contact the following NASA officials:

RETURN to the Table of Contents

- George Lake, University of Washington

The Earth, its relationship to the Sun and Solar System, and the universe in its totality are the domain of the Earth and Space Sciences Project. This effort is employing advanced computers to further our understanding of and ability to predict the dynamically interacting physical, chemical, and biological processes that drive these systems. Its ultimate goal is building an assortment of computer-simulated models that combine complex Earth and space science disciplines.

High-resolution, multidisciplinary models are crucial for their predictive value and for their capacity to estimate beyond what we can measure and observe directly. For example, we cannot ''see'' the beginnings of the universe or even the birth of our own planet, but simulation can provide insight into how they evolved by filling in the gaps left by telescopes or geological records. Current ESS Project investigations include probing the formation of the large-scale universe; modeling the global climate system in the past, present and future; ascertaining the dynamics of the interior of stars; and indexing and searching through massive Earth-observational data sets.

Determining the pertinent interactions, their time scales, and the controls that exist in such systems requires computing power at the highest levels of performance. An objective of the ESS Project is to provide the supercomputers and software tools to facilitate these models. ''Testbed'' facilities allow access to prototype and early-production machines, such as the Convex Exemplar SPP-1 and the MasPar MP-2 at NASA/Goddard Space Flight Center. Other shared testbed facilities are available throughout NASA and at other U.S. government agencies and universities.

Much of the Earth and space sciences relies on data collected from a panoply of satellites and telescopes. There are already massive volumes of data on hand, and one trillion bytes a day will be collected by NASA's Earth Observing System alone. The ESS Project is therefore engaged in developing innovative methods for analysis; these approaches range from visualization and virtual reality to ''intelligent'' information systems and assimilating data into models. Additionally, higher-resolution sensors will require entirely new data retrieval techniques. These endeavors, together with those in modeling, will in turn provide feedback to the system vendors about the effectiveness and limitations of their products, helping them to improve subsequent generations.

If you are interested in additional information on this project or related activities you may access the ESS Home Page on the World Wide Web or you may contact the following NASA officials:

RETURN to the Table of Contents

The NASA IITA component is facilitating and accelerating the implementation of a National Information Infrastructure through NASA science, engineering and technology contributions. This activity is responsive to the Congressional and Presidential goals of building new partnerships between the Federal and non-Federal sectors of U.S. society and has special emphasis on serving new communities.

The IIITA component focuses on four key areas: development of Digital Library Technology; public use of Remote Sensing Data; Aerospace Design and Manufacturing; and, K-12 education over the Internet. Each of these areas supports the development of new technologies to facilitate broader access to NASA data via computer networks.

This NASA activity will foster the development of new and innovative technology to support Digital Libraries; these are libraries that are effectively multimedia digital (electronic) in nature. The focus here is to support the long-term needs of NASA pilot projects already established and for the eventual scale-up to support thousands to millions of users widely distributed over the Internet. Remote Sensing Data is key as this is what will comprise the Digital Libraries. Broad public access to databases of remote sensing images and data over computer networks such as the Internet is also essential; NASA has established a Remote Sensing Public Access Center to manage just such an effort.

NASA is also striving to provide support for Aerospace Design and Manufacturing through ongoing work with aircraft and propulsion companies. This is meant to facilitate the transfer of NASA -developed aerospace design technology to users in major U.S. aerospace companies. NASA is supporting the transfer of sensitive technologies through development of a secure infrastructure for NASA-industry collaborations.

Finally, activities in the area of supporting K-12 education over the Internet will focus on developing curriculum enhancement products for K-12 education, which build on a core program of K-12 education programs at NASA. The result will cause expansion of a broad outreach program to educational product developers in academia and the private sector.

If you are interested in additional information on this project or related activities you may access the IITA Home Page on the World Wide Web or you may contact the following NASA officials:

RETURN to the Table of Contents

- W. Huntress, NASA Headquarters

The Remote Exploration and Experimentation project will develop and demonstrate a space-qualified, spaceborne computing system architecture that requires less than ten watts per billion operations per second. This computing architecture will be scalable from low-powered (sub-watt) systems to higher-powered (hundred-watt) systems that support deep-space missions lasting ten years or more. Deep-space missions require actual (real-time) analysis of sensor data of up to tens of gigabits per second and independent control of complex robotic functions with out intervention from Earth.

This project will: enable and enhance U.S. spaceborne remote sensing and manipulation systems by providing dramatic advances in the performance, reliability and affordability of on-board data processing and control systems; extend U. S. technological leadership in high performance, spaceborne, real-time, durable computing systems and their applications; and, work cooperatively with the U.S. computer industry to assure that NASA technology is commercially available to the U.S. civil, defense and commercial space programs, as well as, for practical, day-to-day applications.

Deep space applications were selected as a primary focus because they have stringent environmental, long-life, and low-power constraints and requirements. Furthermore, long round-trip communications times and low communications bandwidths require on-board data processing and independence from people on Earth. Since near-Earth, airborne, and ground applications are not as mass and power limited, they can use high performance data processing and control systems earlier than deep space missions. Applications that require reliable, real-time responsiveness and that benefit from small size and low power will be addressed by, as well as, gain from this project. NASA will select, in this context, intermediate applications to drive early developments while addressing the primary focus. Some examples of possible applications are: robots for hazardous waste clean-up, search-and-rescue, automated inspection and flexible manufacturing, smart portable atmospheric emission analyzers, remote Earth observing systems with very high resolution instruments, microgravity experiments, and automotive collision avoidance systems.

The Remote Exploration and Experimentation project is currently not active but will resume activities in Fiscal Year 1996.

If you are interested in additional information on this project or related activities you may access the REE Home Page on the World Wide Web or you may contact the following NASA officials:

RETURN to the Table of Contents

NASA HPCC Program Contributions

NASA and its partner agencies are well on their way to achieving high performance computing systems that can operate at a steady rate of at least one trillion arithmetic operations per second -- one teraflop. The Numerical Aerodynamic Simulation (NAS) Parallel Benchmarks were developed to evaluate performance of parallel computing systems for workloads typical in NASA and the aeronautics community and are now in extensive use by several commercial and research communities.

Through a cooperative research agreement with a consortium headed by IBM Corporation, a 160-node SP-2, installed at NASA Ames Research Center, has achieved a 25 fold increase in performance over a Cray Y-MP supercomputer (the fastest supercomputer at the inception of the HPCC Program) on the NAS benchmarks and marked the beginning of the second generation of parallel machines.

A single large high-performance computer has achieved a world record of 143 gigaflops (or 143 billion arithmetic operations per second) on a parallel linear algebra problem. By coupling several large supercomputers over a network, applications exhibiting half a teraflop performance are expected to be demonstrated on the exhibition hall floor at Supercomputing '95 in November, 1995.

The Internet is the creation of the HPCC agencies, but its recent phenomenal growth is the result of educational, public service, and private sector investment. NASA and the Department of Energy (DOE) are accelerating the introduction of new commercial Asynchronous Transfer Mode (ATM) networking technologies through acquisition of experimental 155 Mb/s service at multiple sites in FY 1995. Using the Advanced Communications Technology Satellite, experiments linking computers at 155 Mb/s have been demonstrated earlier this year, with experiments at 622 Mb/s planned for later this year. These experiments should demonstrate the 1997 metric of 100-fold increase in communications capability.

The technologies used in the experiments, coupled with those in support of the National Research and Education Network, lead to high-speed network communications that can be delivered commercially at one-tenth of today's cost of providing the same service.

RETURN to the Table of Contents

Teraflops: What is it??

When discussing the development of new computing technologies, terms like teraflops and gigaflops are spoken as if we should all know what they mean. These are simply units of measurement, which measure the speed with which a computer processes data to perform calculations. Tera means trillion; flops is floating point operations per second. Therefore, teraflops is a trillion floating point operations per second. Teraflops do not exist yet, at least not at sustained rates.

We need to care about ''teraflops''. As you develop high level processing, it trickles down to many applications. A computer that performed in ''teraflops'' could provide farmers with long-range weather predictions and thus, impact U.S. food production. Automobile manufacturers could manipulate huge databases of information instantaneously so they could improve and change designs in real-time, compare present to past, and make predictions. Automobile manufacturers would save design time, which saves money, which keeps the cost of cars down. Cars could have powerful onboard computers with databases of maps, onboard guidance system, and an instrument that tells drivers how much gas it will take to get the nearest gas station, police station, etc.

What we now have is gigaflops computing capability. Giga means billion. This is not good enough because we cannot do all of the things that we need or want to do. We can get a certain amount of data and process it, but not enough to get the information we need. The difference between ''gigaflops'' and ''teraflops'' is represented by the difference between a round trip flight between New York and Boston and a round trip flight between the Earth and the moon.

Now that you are clued in to what teraflops means you may want to ask yourself ''what is a petaflops?''

RETURN to the Table of Contents

Importance of NASA's Role in HPCC

High-performance computing is vital to the Nation's progress in science and engineering. NASA's leadership role in the Federal HPCC program is primarily involved in the development of software to accomplish computational modeling needed in science and engineering.

High-performance computing is critical to strengthening the global competitiveness of the U.S. aeronautics industry.

• NASA's HPCC computational techniques have resulted in engineering productivity improvements that enabled Pratt and Whitney to cut the design time in half for high-pressure jet engine compressors used in the Boeing 777 while reducing fuel consumption, providing savings in both development and operations costs.
• NASA's HPCC Program is working to produce rapid, accurate predictions of the resistance caused by air flowing over an airplane (or drag) to produce superior aircraft designs, reduced certification costs, and improved reliability.

High-performance computing is critical to the vitality of the Earth and space sciences community.

• High-performance computing advances have enabled accurate modeling of the Earth's atmosphere, land surface, and oceans as an important effort in understanding observational data.
• Improved numerical methods and compute power now make possible ocean simulations that more accurately represent ocean structure and lay the foundation for a new coupled atmosphere-ocean climate models that will reduce uncertainties associated with climate change prediction.

The HPCC Program has enabled new models of galaxy and large scale structure formation to be developed and simulated to compare with data from NASA's Great Observatories.

• These new theories are substantially altering our understanding of the formation of stars, galaxies, and the universe.

NASA's Information Infrastructure Technology and Applications Program supports the development of the National Information Infrastructure and provides quality educational tools and curriculum to our nation's children.

• This program has supported four Live from Antarctica events over the Internet and the PBS network, produced the highly received Global Quest video on the using the Internet for education, established six major national Digital Library Testbeds jointly with NSF and the Advanced Research Projects Agency (ARPA), established 26 cooperative agreements and grants for Public Use of Earth and Space Science Data over the Internet, and developed and demonstrated several low cost approaches for establishing Internet connectivity in American K-12 schools.
• The results of NASA's program are now in use in thousands of schools throughout the country.

RETURN to the Table of Contents

Resources: pointers to more HPCC related documentation

Since you have access to the World Wide Web and use Mosaic (or some other Web browser application), we encourage you to go ahead and take a look at more graphic descriptions of NASA's HPCC accomplishments at the following site:

Selected 1995 Program Accomplishments for the NASA HPCC Program: isolated, top-level accomplishments taken from the NASA HPCC 1995 Annual Report

Additional information on the NASA HPCC Program can be accessed from the NASA HPCC Office Home Page.

Information on Federal HPCC objectives and accomplishments are also available in greater detail:

• HPCC Blue Books and Implementation Plans
• Home Page for the National Coordination Office for High Performance Computing and Communications

RETURN to the Table of Contents