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 feasibility of using the Massively Parallel Processor (MPP) and possible future machines with parallel architecture for weather prediction and climate simulation was studied. A shallow water model and a two-level primitive equations model were implemented.
The effect of large amplitude forcing on the growth of a high speed two dimensional jet was investigated experimentally. Two forcing techniques were utilized: mass flow oscillations and a mechanical system. The mass flow oscillation tests were conducted at Strouhal numbers from 0.00052 to 0.045, and peak to peak amplitudes up to 50 percent of the mean exit velocity. The exit Mach number was varied in the range 0.15 to 0.8. The corresponding Reynolds numbers were 8,400 and 45,000. The results indicate no significant change of the jet growth rate or centerline velocity decay compared to the undisturbed free jet. The mechanical forcing system consists of two counter rotating hexagonal cylinders located parallel to the span of the nozzle. Forcing frequencies up to 1,500 Hz were tested. Both symmetric and antisymmetric forcing can be implemented. The results for antisymmetric forcing showed a significant (75 percent) increase of the jet growth rate at an exit Mach number of 0.25 and a Strouhal number of 0.019. At higher rotational speeds, the jet deflected laterally. A deflection angle of 39 deg with respect to the centerline was measured at the maximum rotational speed.
It has long been recognized that simulations form an interesting and important class of computations that may benefit from distributed or parallel processing. Since the point of parallel processing is improved performance, the recent proliferation of multiprocessors requires that we consider the performance issues that naturally arise when attempting to implement a distributed simulation. Three such issues are: (1) the problem of mapping the simulation onto the architecture, (2) the possibilities for performing redundant computation in order to reduce communication, and (3) the avoidance of deadlock due to distributed contention for message-buffer space. These issues are discussed in the context of a battlefield simulation implemented on a medium-scale multiprocessor message-passing architecture.
A procedure is outlined which utilizes parallel processing to solve the inviscid form of the average-passage equation system for multistage turbomachinery along with a description of its implementation in a FORTRAN computer code, MSTAGE. A scheme to reduce the central memory requirements of the program is also detailed. Both the multitasking and I/O routines referred to in this paper are specific to the Cray X-MP line of computers and its associated SSD (Solid-state Storage Device). Results are presented for a simulation of a two-stage rocket engine fuel pump turbine.
A procedure is outlined which utilizes parallel processing to solve the inviscid form of the average-passage equation system for multistage turbomachinery along with a description of its implementation in a FORTRAN computer code, MSTAGE. A scheme to reduce the central memory requirements of the program is also detailed. Both the multitasking and I/O routines referred to are specific to the Cray X-MP line of computers and its associated SSD (Solid-State Disk). Results are presented for a simulation of a two-stage rocket engine fuel pump turbine.
Many traditional algorithms for computing the fast Fourier transform (FFT) on conventional computers are unacceptable for advanced vector and parallel computers because they involve nonunit, power-of-two memory strides. A practical technique for computing the FFT that avoids all such strides and appears to be near-optimal for a variety of current vector and parallel computers is presented. Performance results of a program based on this technique are given. Notable among these results is that a FORTRAN implementation of this algorithm on the CRAY-2 runs up to 77-percent faster than Cray's assembly-coded library routine.
The question of whether multiprocessors should have shared or distributed memory has attracted a great deal of attention. Some researchers argue strongly for building distributed memory machines, while others argue just as strongly for programming shared memory multiprocessors. A great deal of research is underway on both types of parallel systems. Special emphasis is placed on systems with a very large number of processors for computation intensive tasks and considers research and implementation trends. It appears that the two types of systems will likely converge to a common form for large scale multiprocessors.
The objective of this program is to develop generic load models with multiple levels of progressive sophistication to simulate the composite load spectra that are induced in space propulsion system components, representative of Space Shuttle Main Engines (SSME), such as transfer ducts, turbine blades, and liquid oxygen (LOX) posts and system ducting. These models will be developed using two independent approaches. The first approach consists of using state-of-the-art probabilistic methods to describe the individual loading conditions and combinations of these loading conditions to synthesize the composite load spectra simulation. The methodology required to combine the various individual load simulation models (hot-gas dynamic, vibrations, instantaneous position, centrifugal field, etc.) into composite load spectra simulation models will be developed under this program. A computer code incorporating the various individual and composite load spectra models will be developed to construct the specific load model desired. The second approach, which is covered under the options portion of the contract, will consist of developing coupled models for composite load spectra simulation which combine the (deterministic) models for composite load dynamic, acoustic, high-pressure and high rotational speed, etc., load simulation using statistically varying coefficients. These coefficients will then be determined using advanced probabilistic simulation methods with and without strategically selected experimental data. This report covers the efforts of the third year of the contract. The overall program status is that the turbine blade loads have been completed and implemented. The transfer duct loads are defined and are being implemented. The thermal loads for all components are defined and coding is being developed. A dynamic pressure load model is under development. The parallel work on the probabilistic methodology is essentially completed. The overall effort is being integrated in an expert system code specifically developed for this project.
The design and performance of a 0.9 THz to 2.7 THz waveguide tripler are presented. An unusual split block configuration with parallel input and output waveguides accommodates a monolithic membrane diode (MoMeD) circuit. Submicron planar GaAs Schottky diodes in single and antiparallel pairs are implemented with matching filters on a 3-micrometer thick suspended substrate as part of the MoMeD structure. The filters are a combination of short hammerheads and high-low impedance elements. Only a few circuit variations have been measured to date. The best current performance shows an output power of 0.1 microW and an efficiency of 0.002% at the band center frequency of 2.55 THz.
Mesh adaption is a powerful tool for efficient unstructured-grid computations but causes load imbalance among processors on a parallel machine. We present a novel method called PLUM to dynamically balance the processor workloads with a global view. This paper presents the implementation and integration of all major components within our dynamic load balancing strategy for adaptive grid calculations. Mesh adaption, repartitioning, processor assignment, and remapping are critical components of the framework that must be accomplished rapidly and efficiently so as not to cause a significant overhead to the numerical simulation. A data redistribution model is also presented that predicts the remapping cost on the SP2. This model is required to determine whether the gain from a balanced workload distribution offsets the cost of data movement. Results presented in this paper demonstrate that PLUM is an effective dynamic load balancing strategy which remains viable on a large number of processors.
We have described a challenging neural-network hardware implementation that would mate a 64x64-pixel infrared sensor directly on to a 3-dimensionally packaged set of neural processing chips with parallel input for high speed image processing. Non linearity effects and range compression are also discussed.
The development of new mission concepts requires efficient methodologies to analyze, design and simulate the concepts before implementation. New mission concepts are increasingly considering the use of ion thrusters for fuel-efficient navigation in deep space. This paper presents parallel, evolutionary computing methods to design trajectories of spacecraft propelled by ion thrusters and to assess the trade-off between delivered payload mass and required flight time. The developed methods utilize a distributed computing environment in order to speed up computation, and use evolutionary algorithms to find globally Pareto-optimal solutions. The methods are coupled with two main traditional trajectory design approaches, which are called direct and indirect. In the direct approach, thrust control is discretized in either arc time or arc length, and the resulting discrete thrust vectors are optimized. In the indirect approach, a thrust control problem is transformed into a costate control problem, and the initial values of the costate vector are optimized. The developed methods are applied to two problems: 1) an orbit transfer around the Earth and 2) a transfer between two distance retrograde orbits around Europa, the closest to Jupiter of the icy Galilean moons. The optimal solutions found with the present methods are comparable to other state-of-the-art trajectory optimizers and to analytical approximations for optimal transfers, while the required computational time is several orders of magnitude shorter than other optimizers thanks to an intelligent design of control vector discretization, advanced algorithmic parameterization, and parallel computing.
The Geostationary Lightning Mapper (GLM) is a single channel, near-IR imager/optical transient event detector, used to detect, locate and measure total lightning activity over the full-disk as part of a 3-axis stabilized, geostationary weather satellite system. The next generation NOAA Geostationary Operational Environmental Satellite (GOES-R) series with a planned launch readiness in December 2014 will carry a GLM that will provide continuous day and night observations of lightning from the west coast of Africa (GOES-E) to New Zealand (GOES-W) when the constellation is fUlly operational. The mission objectives for the GLM are to 1) provide continuous, full-disk lightning measurements for storm warning and nowcasting, 2) provide early warning of tornadic activity, and 3) accumulate a long-term database to track decadal changes of lightning. The GLM owes its heritage to the NASA Lightning Imaging Sensor (1997-Present) and the Optical Transient Detector (1995-2000), which were developed for the Earth Observing System and have produced a combined 13 year data record of global lightning activity. Instrument formulation studies were completed in March 2007 and the implementation phase to develop a prototype model and up to four flight models will be underway in the latter part of 2007. In parallel with the instrument development, a GOES-R Risk Reduction Team and Algorithm Working Group Lightning Applications Team have begun to develop the Level 2 algorithms and applications. Proxy total lightning data from the NASA Lightning Imaging Sensor on the Tropical Rainfall Measuring Mission (TRMM) satellite and regional test beds (e.g., Lightning Mapping Arrays in North Alabama and the Washington DC Metropolitan area) are being used to develop the pre-launch algorithms and applications, and also improve our knowledge of thunderstorm initiation and evolution. Real time lightning mapping data are being provided in an experimental mode to selected National Weather Service (NWS) forecast offices in Southern and Eastern Region. This effort is designed to help improve our understanding of the application of these data in operational settings.
Analog computational circuits have been demonstrated to provide substantial improvements in power and speed relative to digital circuits, especially for applications requiring extreme parallelism but only modest precision. Deep machine learning is one such area and stands to benefit greatly from analog and mixed-signal implementations. However, even at modest precisions, offsets and non-linearity can degrade system performance. Furthermore, in all but the simplest systems, it is impossible to directly measure the intermediate outputs of all sub-circuits. The result is that circuit designers are unable to accurately evaluate the non-idealities of computational circuits in-situ and are therefore unable to fully utilize measurement results to improve future designs. In this paper we present a technique to use deep learning frameworks to model physical systems. Recently developed libraries like TensorFlow make it possible to use back propagation to learn parameters in the context of modeling circuit behavior. Offsets and scaling errors can be discovered even for sub-circuits that are deeply embedded in a computational system and not directly observable. The learned parameters can be used to refine simulation methods or to identify appropriate compensation strategies. We demonstrate the framework using a mixed-signal convolution operator as an example circuit.
Many applications require to learn, mine, analyze and visualize large-scale graphs. These graphs are often too large to be addressed efficiently using conventional graph processing technologies. Fortunately, recent research efforts find out graph sampling and random walk, which significantly reduce the size of original graphs, can benefit the tasks of learning, mining, analyzing and visualizing large graphs by capturing the desirable graph properties. This paper introduces C-SAW, the first framework that accelerates Sampling and Random Walk framework on GPUs. Particularly, C-SAW makes three contributions: First, our framework provides a generic API which allows users to implement a wide range of sampling and random walk algorithms with ease. Second, offloading this framework on GPU, we introduce warp-centric parallel selection, and two novel optimizations for collision migration. Third, towards supporting graphs that exceed the GPU memory capacity, we introduce efficient data transfer optimizations for out-of-memory and multi-GPU sampling, such as workload-aware scheduling and batched multi-instance sampling. Taken together, our framework constantly outperforms the state of the art projects in addition to the capability of supporting a wide range of sampling and random walk algorithms.
Rendezvous algorithms encode a communication pattern that is useful when processors sending data do not know who the receiving processors should be, or vice versa. The idea is to define an intermediate decomposition where datums from different sending processors can ”rendezvous” to perform a computation, in a manner that both the senders and eventual receivers of the results can identify the appropriate rendezvous processor. Though they were originally designed for interpolating between overlaid grids with independent parallel decompositions (Plimpton et al., 2004), we have recently found rendezvous algorithms useful for a variety of operations in particle- or grid-based simulation codes when running large problems on large numbers of processors. In particular, we show they can perform well when a load-balanced intermediate decomposition is randomized and not spatial, requiring all-to-all communication to move data between processors. In this case rendezvous algorithms leverage the large bisection communication bandwidths which parallel machines provide. We describe how rendezvous algorithms work in a scientific computing context and give specific examples for molecular dynamics and Direct Simulation Monte Carlo codes which result in dramatic performance improvements versus simpler algorithms which do not scale as well. We explain how a generic rendezvous algorithm can be implemented, and also point out similarities with the MapReduce paradigm popularized by Google and Hadoop.
In this work we present a 2D distributed memory algorithm for overlap detection and introduce a novel design and implementation for layout simplification where we perform a transitive reduction from overlap graph to string graph in the context of de novo genome assembly.
Parallel-beam tomography systems at synchrotron facilities have limited field of view (FOV) determined by the available beam size and detector system coverage. Scanning the full size of samples bigger than the FOV requires various data acquisition schemes such as grid scan, 360-degree scan with offset center-of-rotation (COR), helical scan, or combinations of these schemes. Though straightforward to implement, these scanning techniques have not often been used due to the lack of software and methods to process such types of data in an easy and automated fashion. The ease of use and automation is critical at synchrotron facilities where using visual inspection in data processing steps such as image stitching, COR determination, or helical data conversion is impractical due to the large size of datasets. Here, we provide methods and their implementations in a Python package, named Algotom, for not only processing such data types but also with the highest quality possible. The efficiency and ease of use of these tools can help to extend applications of parallel-beam tomography systems.
Refine your search to explore more results.