SEARCH · Search NASA
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
The group's CCN counter is described. It is a static, horizontal, parallel plate thermal gradient diffusion chamber. Examples of the application of the CCN are presented and include the CCN spectra measured during the winter of 1978-79 near Elk Mountain, Wyoming. Comparisons of droplet concentrations derived from upwind CCN spectra are covered.
The paper discusses software reliability as it applies particularly to design and evaluation of flight-critical digital avionics systems. Measures of software reliability, measurement methods and reliability (macro-) models are discussed. Recent work assessing their accuracy in predicting software errors in 'fly-by-wire' Newtonian applications is presented. Additional, detailed topics are discussed including software error distributions (e.g. catastrophic vs. noncatastrophic) and the effects of system growth/maturity on reliability improvement. In practical flight-critical digital applications, software reliability improvement is sought through use of parallel, redundant software (i.e. N-version programming) or backup software that can be invoked in the event of (primary) software failure. Achievable reliability levels are however highly sensitive to common-mode specification and programming errors. Recent data correlating these errors with net software reliability are discussed.
Solution techniques for handling large scale engineering optimization problems are reviewed. Potentials for practical applications as well as their limited capabilities are discussed. A new solution algorithm for design sensitivity is proposed. The algorithm is based upon the multilevel substructuring concept to be coupled with the adjoint method of sensitivity analysis. There are no approximations involved in the present algorithm except the usual approximations introduced due to the discretization of the finite element model. Results from the six- and thirty-bar planar truss problems show that the proposed multilevel scheme for sensitivity analysis is more effective (in terms of computer incore memory and the total CPU time) than a conventional (one level) scheme even on small problems. The new algorithm is expected to perform better for larger problems and its applications on the new generation of computer hardwares with 'parallel processing' capability is very promising.
There is an evident need to discover a means of establishing reliable, implementable controls for systems that are plagued by nonlinear and, or uncertain, model dynamics. The development of a generic controller design tool for tough-to-control systems is reported. The method utilizes a moving grid, time infinite element based solution of the necessary conditions that describe an optimal controller for a system. The technique produces a discrete feedback controller. Real time laboratory experiments are now being conducted to demonstrate the viability of the method. The algorithm that results is being implemented in a microprocessor environment. Critical computational tasks are accomplished using a low cost, on-board, multiprocessor (INMOS T800 Transputers) and parallel processing. Progress to date validates the methodology presented. Applications of the technique to the control of highly flexible robotic appendages are suggested.
The use of hybrid interconnection structures that combine link connections and bus connections for real-time parallel processing is discussed. Idealistic parallel computation models for two real-time computing applications are described with attention given to a tightly coupled network model for object tracking and a network model for image processing. Consideration is given to the following different interconnection structures: the crossbar, the hypercube, the circular linked array, and the bus array.
Recent developments at the NASA AMES Research Center's NAS Division have demonstrated that the new generation of NUMA based Symmetric Multi-Processing systems (SMPs), such as the Silicon Graphics Origin 2000, can successfully execute legacy vector oriented CFD production codes at sustained rates far exceeding processing rates possible on dedicated 16 CPU Cray C90 systems. This high level of performance is achieved via shared memory based Multi-Level Parallelism (MLP). This programming approach, developed at NAS and outlined below, is distinct from the message passing paradigm of MPI. It offers parallelism at both the fine and coarse grained level, with communication latencies that are approximately 50-100 times lower than typical MPI implementations on the same platform. Such latency reductions offer the promise of performance scaling to very large CPU counts. The method draws on, but is also distinct from, the newly defined OpenMP specification, which uses compiler directives to support a limited subset of multi-level parallel operations. The NAS MLP method is general, and applicable to a large class of NASA CFD codes.
Historically NASA has trained teams of astronauts by bringing them to the Johnson Space Center in Houston to undergo generic training, followed by mission-specific training. This latter training begins after a crew has been selected for a mission (perhaps two years before the launch of that mission). While some Space Shuttle flights have included an astronaut from a foreign country, the International Space Station will be consistently crewed by teams comprised of astronauts from two or more of the partner nations. The cost of training these international teams continues to grow in both monetary and personal terms. Thus, NASA has been seeking alternative training approaches for the International Space Station program. Since 1994 we have been developing, testing, and refining shared virtual environments for astronaut team training, including the use of virtual environments for use while in or in transit to the task location. In parallel with this effort, we have also been preparing applications for training teams of military personnel engaged in peacekeeping missions. This paper will describe the applications developed to date, some of the technological challenges that have been overcome in their development, and the research performed to guide the development and to measure the efficacy of these shared environments as training tools.
We study data traffic on distributed shared memory machines and conclude that data placement and grouping improve performance of scientific codes. We present several methods which user can employ to improve data traffic in his code. We report on implementation of a tool which detects the code fragments causing data congestions and advises user on improvements of data routing in these fragments. The capabilities of the tool include deduction of data alignment and affinity from the source code; detection of the code constructs having abnormally high cache or TLB misses; generation of data placement constructs. We demonstrate the capabilities of the tool on experiments with NAS parallel benchmarks and with a simple computational fluid dynamics application ARC3D.
Column-parallel analog-to-digital converters (ADCs) for imagers involve simultaneous operation of many ADCs. Single-slope ADCs are well adapted to this use because of their simplicity. Each ADC contains a comparator, comparing its input signal level to an increasing reference signal (ramp). When the ramp is equal to the input, the comparator triggers a latch that captures an encoded counter value (code). Knowing the captured code, the ramp value and hence the input signal are determined. In a column-parallel ADC, each column contains only the comparator and the latches; the ramp and code generation are shared. In conventional latch or flip-flop circuits, there is an input stage that tracks the input signal, and this stage consumes switching current every time the input changes. With many columns, many bits, and high code rates, this switching current can be substantial. It will also generate noise that may corrupt the analog signals. A latch was designed that does not track the input, and consumes power only at the instant of latching the data value. The circuit consists of two S-R (set-reset) latches, gated by the comparator. One is set by high data values and the other by low data values. The latches are cross-coupled so that the first one to set blocks the other. In order that the input data not need an inversion, which would consume power, the two latches are made in complementary polarity. This requires complementary gates from the comparator, instead of complementary data values, but the comparator only triggers once per conversion, and usually has complementary outputs to begin with. An efficient CMOS (complementary metal oxide semiconductor) implementation of this circuit is shown in the figure, where C is the comparator output, D is the data (code), and Q0 and Q1 are the outputs indicating the capture of a zero or one value. The latch for Q0 has a negative-true set signal and output, and is implemented using OR-AND-INVERT logic, while the latch for Q1 uses positive- true signals and is implemented using AND-OR-INVERT logic. In this implementation, both latches are cleared when the comparator is reset. Two redundant transistors are removed from the reset side of each latch, making for a compact layout. CMOS imagers with column-parallel ADCs have demonstrated high performance for remote sensing applications. With this latch circuit, the power consumption and noise can be further reduced. This innovation can be used in CMOS imagers and very-low-power electronics
GridPACK is a software framework consisting of a set of modules designed to simplify the development of programs that model the power grid and run on parallel, high performance computing platforms. It also contains several fully developed applications, including powerflow, dynamic simulation, state estimation, Kalman filter analysis (dynamic state estimation), contingency analysis and real time path rating. These applications can be used either standalone or as components in more complicated workflows that combine several different types of application together. The framework modules are available as a combination of libraries and software templates and consist of components for setting up and distributing power grid networks, support for modeling the behavior of individual buses and branches in the network, converting the network models to the corresponding algebraic equations, and parallel routines for manipulating and solving large algebraic systems. The framework also contains a module for distributing tasks evenly amongst computing resources, even if individual tasks vary widely in their execution times. Additional modules support input and output, basic statistical analysis of contingency based calculations, distributed data structures, as well as basic profiling and error management.
In this tutorial, we will discuss top five current RISC microprocessors: The IBM Power2, which is used in the IBM RS6000/590 workstation and in the IBM SP2 parallel supercomputer, the DEC Alpha, which is in the DEC Alpha workstation and in the Cray T3D; the MIPS R8000, which is used in the SGI Power Challenge; the HP PA-RISC 7100, which is used in the HP 700 series workstations and in the Convex Exemplar; and the Cray proprietary processor, which is used in the new Cray J916. The architecture of these microprocessors will first be presented. The effective performance of these processors will then be compared, both by citing standard benchmarks and also in the context of implementing a real applications. In the process, different programming models such as data parallel (CM Fortran and HPF) and message passing (PVM and MPI) will be introduced and compared. The latest NAS Parallel Benchmark (NPB) absolute performance and performance per dollar figures will be presented. The next generation of the NP13 will also be described. The tutorial will conclude with a discussion of general trends in the field of high performance computing, including likely future developments in hardware and software technology, and the relative roles of vector supercomputers tightly coupled parallel computers, and clusters of workstations. This tutorial will provide a unique cross-machine comparison not available elsewhere.
Classical mesh partitioning algorithms were designed for rather static situations, and their straightforward application in a dynamical framework may lead to unsatisfactory results, e.g., excessive data migration among processors. Furthermore, special attention should be paid to their amenability to parallelization. In this paper, a novel parallel method for the dynamic partitioning of adaptive unstructured meshes is described. It is based on a linear representation of the mesh using self-avoiding walks.
A reduced-thickness backlighter for an autostereoscopic display is disclosed having a lightguide and at least one light source parallel to an edge of the lightguide so as to be substantially coplanar with the lightguide. The lightguide is provided with a first surface which has a plurality of reflective linear regions, such as elongated grooves or glossy lines, parallel to the illuminated edge of the lightguide. Preferably the lightguide further has a second surface which has a plurality of lenticular lenses for reimaging the reflected light from the linear regions into a series of thin vertical lines outside the guide. Because of the reduced thickness of the backlighter system, autostereoscopic viewing is enabled in applications requiring thin backlighter systems. In addition to taking up less space, the reduced-thickness backlighter uses less lamps and less power. For accommodating 2-D applications, a 2-D diffuser plate or a 2-D lightguide parallel to the 3-D backlighter is disclosed for switching back and forth between 3-D viewing and 2-D viewing.
There has recently been much effort behind the development of NDE methods applicable to the detection of surface/subsurface flaws in thin metallic structures with a rapid scan capability. One such method, an electromagnetic technique using a current-sheet parallel to the surface of a specimen in order to induce eddy current flow shows a high potential for satisfying the rapid scan requirement stated above. The technique is based on the detection of flaw-induced magnetic field components normal to the specimen surface by an appropriate detection mechanism positioned above the current-sheet as shown schematically in Fig. 1. As indicated in this figure, the current-sheet separates the source of the normal magnetic field components from the detector in such a way that the electric and magnetic properties of the current-sheet can be a major factor affecting the strength of the detected signals. The purpose of the present study is, therefore, to perform a detailed investigation on the effect of the material properties of the current-sheet on the detected signal strength and to establish a simple theoretical model for the detection mechanism.
This paper presents a detailed performance analysis of a multi-block overset grid compu- tational fluid dynamics app!ication on multiple state-of-the-art computer architectures. The application is implemented using a hybrid MPI+OpenMP programming paradigm that exploits both coarse and fine-grain parallelism; the former via MPI message passing and the latter via OpenMP directives. The hybrid model also extends the applicability of multi-block programs to large clusters of SNIP nodes by overcoming the restriction that the number of processors be less than the number of grid blocks. A key kernel of the application, namely the LU-SGS linear solver, had to be modified to enhance the performance of the hybrid approach on the target machines. Investigations were conducted on cacheless Cray SX6 vector processors, cache-based IBM Power3 and Power4 architectures, and single system image SGI Origin3000 platforms. Overall results for complex vortex dynamics simulations demonstrate that the SX6 achieves the highest performance and outperforms the RISC-based architectures; however, the best scaling performance was achieved on the Power3.
TeraChem was born in 2008 with the goal of providing fast on-the-fly electronic structure calculations to facilitate ab initio molecular dynamics studies of large biochemical systems such as photoswitchable proteins and multichromophoric antenna complexes. Originally developed for videogaming applications, graphics processing units (GPUs) offered a low-cost parallel computer architecture that became more accessible for general-purpose GPU computing with the release of CUDA in 2007. The evaluation of the electron repulsion integrals (ERIs) is a major bottleneck in electronic structure codes and provides an attractive target for acceleration on GPUs. Thus, highly efficient routines for evaluation of and contractions between the ERIs and density matrices were implemented in TeraChem. Here, electronic structure methods were developed and implemented to leverage these integral contraction routines, resulting in the first quantum chemistry package designed from the ground up for GPUs. This GPU acceleration makes TeraChem capable of performing large-scale ground and excited state calculations in the gas and condensed phase. Today, TeraChem's speed forms the basis for a suite of quantum chemistry applications, including optimization and dynamics of proteins, automated and interactive chemical discovery tools, and large-scale nonadiabatic dynamics simulations.
Hardware and software for Merlin, a new kind of massively parallel computing system, are described. Eight computers are linked as a 300-MIPS prototype to develop system software for a larger Merlin network with 16 to 64 nodes, totaling 600 to 3000 MIPS. These working prototypes help refine a mapped reflective memory technique that offers a new, very general way of linking many types of computer to form supercomputers. Processors share data selectively and rapidly on a word-by-word basis. Fast firmware virtual circuits are reconfigured to match topological needs of individual application programs. Merlin's low-latency memory-sharing interfaces solve many problems in the design of high-performance computing systems. The Merlin prototypes are intended to run parallel programs for scientific applications and to determine hardware and software needs for a future Teraflops Merlin network.
This paper presents an algorithm to find the correspondences of points representing dominant feature in robot stereo vision. The algorithm consists of two main steps: dominant point extraction and dominant point matching. In the feature extraction phase, the algorithm utilizes the widely used Moravec Interest Operator and two other operators: the Prewitt Operator and a new operator called Gradient Angle Variance Operator. The Interest Operator in the Moravec algorithm was used to exclude featureless areas and simple edges which are oriented in the vertical, horizontal, and two diagonals. It was incorrectly detecting points on edges which are not on the four main directions (vertical, horizontal, and two diagonals). The new algorithm uses the Prewitt operator to exclude featureless areas, so that the Interest Operator is applied only on the edges to exclude simple edges and to leave interesting points. This modification speeds-up the extraction process by approximately 5 times. The Gradient Angle Variance (GAV), an operator which calculates the variance of the gradient angle in a window around the point under concern, is then applied on the interesting points to exclude the redundant ones and leave the actual dominant ones. The matching phase is performed after the extraction of the dominant points in both stereo images. The matching starts with dominant points in the left image and does a local search, looking for corresponding dominant points in the right image. The search is geometrically constrained the epipolar line of the parallel-axes stereo geometry and the maximum disparity of the application environment. If one dominant point in the right image lies in the search areas, then it is the corresponding point of the reference dominant point in the left image. A parameter provided by the GAV is thresholded and used as a rough similarity measure to select the corresponding dominant point if there is more than one point the search area. The correlation is used as a final decision tool when there is still more than one point in the search area. If there is no dominant point in the search area of if the points in the search area are below a correlation threshold, then the dominant point in the reference image is occluded and can not be corresponded. The algorithm has been modeled, implemented and shown to be fast, robust and parallel. The parallelism is created from three main features: locality of the operators; a memory optimization scheme; and the ability to fully parallelize the extraction phase which is the most computational intensive task in the algorithm. The last feature is achieved by performing the extraction phase on the two images simultaneously.