|
Technologies for the 21st Century
HECC |
 
|
|
|
|
|||
 |
|
||||||||
|---|---|---|---|---|---|---|---|---|---|
|
|
||||||||
|
|
|||||||||
Key research and development (R&D) areas
|
HECC R&D investments provide the foundation for U.S. leadership in high end computing. HECC research explores the advanced concepts in quantum, biological, and optical computing that will keep the U.S. in the forefront of computing breakthroughs for years to come through its hardware and software innovations; algorithms for physical, chemical, and biological modeling and simulation of these processes in complex systems; and information-intensive science and engineering applications. HECC R&D targets four areas:
Federal investments in all four of these areas will enable development of the distributed, computation-intensive applications necessary to support future U.S. science and engineering research, economic competitiveness, and national security priorities. |
||||||||
|
|
|||||||||
|
HECC goals:
|
The medium range (three to five year) technology development goal for HECC R&D is the achievement of major improvements in usability and effectiveness in teraflops-scale systems across a wide range of applications. Longer range goals (more than five years) include a firmer understanding of the device technology, algorithms, and software that are required for petaflops-level computation and exabyte-level mass storage. |
||||||||
|
|
|||||||||
|
1. System software
|
HECC system software R&D will focus on developing parallel software tools, operating systems, program development environments, and performance monitoring for use in distributed, scalable systems. These tools and system software will improve scalability, throughput, speedup, portability, and programmability. HECC R&D will support parallel systems software such as innovative languages and their compilers, debuggers, performance monitors, scalable operating systems and I/O, program development environments, scientific visualization, data management, as well as frameworks to link software components. Large scale data management will also require the development and adoption of new storage technologies ranging from exabyte tape robots to large, fast, on-chip memory. |
||||||||
|
|
|||||||||
|
2. Future generations computing research
|
HECC will support research and technology necessary to achieve petaflops-level computation and exabyte-level mass storage by developing innovative technology for software, hardware, architecture, and components. Research based on the shared-memory programming model will create techniques to overcome memory latency through multithreading, better caching algorithms, or other means. R&D will also focus on software to support these techniques, applications design studies, architecture point design studies, and new approaches to component technology. Also included in this area is research on transportable software technologies that scale the symmetric multiprocessor systems currently available (with several to hundreds of processors) to systems with very large numbers (tens of thousands) of processors. Concepts in the research stage include logic circuits based on current semiconducting materials as well as novel materials, such as low and high temperature superconductors, and quantum mechanics-based devices, such as rapid single-flux quantum devices used for logic circuits and memories. Research will be conducted on the technological, algorithmic, and architectural foundations for amorphous computing with programmable materials. Other promising research concepts are on- and off-chip interconnections based on guided optics that employ wavelength division multiplexing, and massive holographic optical memory devices for mass storage. Basic research will be conducted on innovative logic devices based on nanotechnology and biological materials that may exploit the information contained in large molecules such as deoxyribonucleic acid (DNA). |
||||||||
|
|
|||||||||
|
3. Agency applications
|
R&D will focus on incorporating HECC technologies into agency applications, as well as on developing high performance computational science and its underlying algorithms to ensure that key applications will run at their full potential. Many Computing, Information, and Communications (CIC) agencies support scientific mission-driven applications projects requiring large scale computation-intensive or data-intensive operations. These projects span the spectrum of scientific problems from cosmology to global climate modeling to short range weather prediction to protein folding to quantum chromodynamics. R&D is required to support advances in fast, efficient algorithms for computational sciences addressing emerging computational challenges, including very large sparse matrix-based problems, searching, sorting, and pattern matching. Research on algorithms with large amounts of concurrency, fault tolerance, and latency hiding is also crucial to the use of the high end computational systems of the future. |
||||||||
|
|
|||||||||
|
4. Infrastructure for research
|
By planning and coordinating the design, implementation, and maintenance of a state-of-the-art computational infrastructure, HECC R&D will ensure that the full potential of computational research facilities, large scale test systems, and high performance networks is realized. Information exchange and coordination will ensure the continued development of a balanced infrastructure with the maximum computational strength and network bandwidth such as that available through the CIC-supported centers. This infrastructure is critical in order to continue ongoing research in large scale computation-intensive problems. |
||||||||
|
|
|||||||||
|
HECC R&D examples: |
Examples of HECC R&D are described in the rest of this section. |
||||||||
|
|
|
||||||||
|
|
|||||||||
