Search NASA⌕ Search

SEARCH · Search NASA

Results for “Parallel processors”

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.

At least 163 records · Page 9

Speeding up parallel processing

In 1967 Amdahl expressed doubts about the ultimate utility of multiprocessors. The formulation, now called Amdahl's law, became part of the computing folklore and has inspired much skepticism about the ability of the current generation of massively parallel processors to efficiently deliver all their computing power to programs. The widely publicized recent results of a group at Sandia National Laboratory, which showed speedup on a 1024 node hypercube of over 500 for three fixed size problems and over 1000 for three scalable problems, have convincingly challenged this bit of folklore and have given new impetus to parallel scientific computing.

Denning, Peter J.↗

Image segmentation by iterative parallel region growing and splitting

The spatially constrained clustering (SCC) iterative parallel region-growing technique is applied to image analysis. The SCC algorithm is implemented on the massively parallel processor at NASA Goddard. Most previous region-growing approaches have the drawback that the segmentation produced depends on the order in which portions of the image are processed. The ideal solution to this problem (merging only the single most similar pair of spatially adjacent regions in the image in each iteration) becomes impractical except for very small images, even on a massively parallel computer. The SCC algorithm overcomes these problems by performing, in parallel, the best merge within each of a set of local, possibly overlapping, subimages. A region-splitting stage is also incorporated into the algorithm, but experiments show that region splitting generally does not improve segmentation results. The SCC algorithm has been tested on various imagery data, and test results for a Landsat TM image are summarized.

Tilton, James C.↗

Large Scale Finite Element Modeling Using Scalable Parallel Processing

An iterative solver for use with finite element codes was developed for the Cray T3D massively parallel processor at the Jet Propulsion Laboratory. Finite element modeling is useful for simulating scattered or radiated electromagnetic fields from complex three-dimensional objects with geometry variations smaller than an electrical wavelength.

finite element modeling parallel processing iterat↗

Visualization of unsteady computational fluid dynamics

The current computing environment that most researchers are using for the calculation of 3D unsteady Computational Fluid Dynamic (CFD) results is a super-computer class machine. The Massively Parallel Processors (MPP's) such as the 160 node IBM SP2 at NAS and clusters of workstations acting as a single MPP (like NAS's SGI Power-Challenge array) provide the required computation bandwidth for CFD calculations of transient problems. Work is in progress on a set of software tools designed specifically to address visualizing 3D unsteady CFD results in these super-computer-like environments. The visualization is concurrently executed with the CFD solver. The parallel version of Visual3, pV3 required splitting up the unsteady visualization task to allow execution across a network of workstation(s) and compute servers. In this computing model, the network is almost always the bottleneck so much of the effort involved techniques to reduce the size of the data transferred between machines.

Haimes, Robert↗

Parallel VLSI architecture emulation and the organization of APSA/MPP

The Applicative Programming System Architecture (APSA) combines an applicative language interpreter with a novel parallel computer architecture that is well suited for Very Large Scale Integration (VLSI) implementation. The Massively Parallel Processor (MPP) can simulate VLSI circuits by allocating one processing element in its square array to an area on a square VLSI chip. As long as there are not too many long data paths, the MPP can simulate a VLSI clock cycle very rapidly. The APSA circuit contains a binary tree with a few long paths and many short ones. A skewed H-tree layout allows every processing element to simulate a leaf cell and up to four tree nodes, with no loss in parallelism. Emulation of a key APSA algorithm on the MPP resulted in performance 16,000 times faster than a Vax. This speed will make it possible for the APSA language interpreter to run fast enough to support research in parallel list processing algorithms.

Odonnell, John T.↗

Parallel algorithms for interactive manipulation of digital terrain models

Interactive three-dimensional graphics applications, such as terrain data representation and manipulation, require extensive arithmetic processing. Massively parallel machines are attractive for this application since they offer high computational rates, and grid connected architectures provide a natural mapping for grid based terrain models. Presented here are algorithms for data movement on the massive parallel processor (MPP) in support of pan and zoom functions over large data grids. It is an extension of earlier work that demonstrated real-time performance of graphics functions on grids that were equal in size to the physical dimensions of the MPP. When the dimensions of a data grid exceed the processing array size, data is packed in the array memory. Windows of the total data grid are interactively selected for processing. Movement of packed data is needed to distribute items across the array for efficient parallel processing. Execution time for data movement was found to exceed that for arithmetic aspects of graphics functions. Performance figures are given for routines written in MPP Pascal.

Davis, E. W.↗

The factorization of large composite numbers on the MPP

The continued fraction method for factoring large integers (CFRAC) was an ideal algorithm to be implemented on a massively parallel computer such as the Massively Parallel Processor (MPP). After much effort, the first 60 digit number was factored on the MPP using about 6 1/2 hours of array time. Although this result added about 10 digits to the size number that could be factored using CFRAC on a serial machine, it was already badly beaten by the implementation of Davis and Holdridge on the CRAY-1 using the quadratic sieve, an algorithm which is clearly superior to CFRAC for large numbers. An algorithm is illustrated which is ideally suited to the single instruction multiple data (SIMD) massively parallel architecture and some of the modifications which were needed in order to make the parallel implementation effective and efficient are described.

Mckurdy, Kathy J.↗

Visualization of unsteady computational fluid dynamics

A brief summary of the computer environment used for calculating three dimensional unsteady Computational Fluid Dynamic (CFD) results is presented. This environment requires a super computer as well as massively parallel processors (MPP's) and clusters of workstations acting as a single MPP (by concurrently working on the same task) provide the required computational bandwidth for CFD calculations of transient problems. The cluster of reduced instruction set computers (RISC) is a recent advent based on the low cost and high performance that workstation vendors provide. The cluster, with the proper software can act as a multiple instruction/multiple data (MIMD) machine. A new set of software tools is being designed specifically to address visualizing 3D unsteady CFD results in these environments. Three user's manuals for the parallel version of Visual3, pV3, revision 1.00 make up the bulk of this report.

Haimes, Robert↗

Analysis of a hardware and software fault tolerant processor for critical applications

Computer systems for critical applications must be designed to tolerate software faults as well as hardware faults. A unified approach to tolerating hardware and software faults is characterized by classifying faults in terms of duration (transient or permanent) rather than source (hardware or software). Errors arising from transient faults can be handled through masking or voting, but errors arising from permanent faults require system reconfiguration to bypass the failed component. Most errors which are caused by software faults can be considered transient, in that they are input-dependent. Software faults are triggered by a particular set of inputs. Quantitative dependability analysis of systems which exhibit a unified approach to fault tolerance can be performed by a hierarchical combination of fault tree and Markov models. A methodology for analyzing hardware and software fault tolerant systems is applied to the analysis of a hypothetical system, loosely based on the Fault Tolerant Parallel Processor. The models consider both transient and permanent faults, hardware and software faults, independent and related software faults, automatic recovery, and reconfiguration.

Dugan, Joanne B.↗

An improved multilevel optimization approach for the design of complex engineering systems

Multilevel optimization methods are being considered for the design of complex systems on distributed networks of computers or even parallel processors. An obstacle to the use of multilevel methods is that they can be computationally expensive because of the cycling necessary to account for the coupling between the subproblems. This research effort aims at increasing the efficiency of multilevel optimization by adapting two techniques that are widely used in conventional one-level optimization: constraint approximation and temporary constraint deletion. These improvements are implemented and tested on three-, ten- and 52-bar planar truss designs. The results show that for larger problems (approximately 100 design variables and larger), the cost of analysis dominates the total cost so that multilevel optimization is no more expensive than one-level optimization. If parallel processing is used or the analysis process itself is decomposed, then multilevel optimization stands to become more economical than one-level optimization.

Barthelemy, J.-F. M.↗

Visualization of Unsteady Computational Fluid Dynamics

The current compute environment that most researchers are using for the calculation of 3D unsteady Computational Fluid Dynamic (CFD) results is a super-computer class machine. The Massively Parallel Processors (MPP's) such as the 160 node IBM SP2 at NAS and clusters of workstations acting as a single MPP (like NAS's SGI Power-Challenge array and the J90 cluster) provide the required computation bandwidth for CFD calculations of transient problems. If we follow the traditional computational analysis steps for CFD (and we wish to construct an interactive visualizer) we need to be aware of the following: (1) Disk space requirements. A single snap-shot must contain at least the values (primitive variables) stored at the appropriate locations within the mesh. For most simple 3D Euler solvers that means 5 floating point words. Navier-Stokes solutions with turbulence models may contain 7 state-variables. (2) Disk speed vs. Computational speeds. The time required to read the complete solution of a saved time frame from disk is now longer than the compute time for a set number of iterations from an explicit solver. Depending, on the hardware and solver an iteration of an implicit code may also take less time than reading the solution from disk. If one examines the performance improvements in the last decade or two, it is easy to see that depending on disk performance (vs. CPU improvement) may not be the best method for enhancing interactivity. (3) Cluster and Parallel Machine I/O problems. Disk access time is much worse within current parallel machines and cluster of workstations that are acting in concert to solve a single problem. In this case we are not trying to read the volume of data, but are running the solver and the solver outputs the solution. These traditional network interfaces must be used for the file system. (4) Numerics of particle traces. Most visualization tools can work upon a single snap shot of the data but some visualization tools for transient problems require dealing with time.

Haimes, Robert↗

Application of the multigrid method to grid generation

The multigrid method (MGM), used to numerically solve the pair of nonlinear elliptic equations commonly used to generate two dimensional boundary-fitted coordinate systems is discussed. Two different geometries are considered: one involving a coordinate system fitted about a circle and the other selected for an impinging jet flow problem. Two different relaxation schemes are tried: one is successive point overrelaxation and the other is a four-color scheme vectorizeable to take advantage of a parallel processor computer for greater computational speed. Results using MGM are compared with those using SOR (doing successive overrelaxations with the corresponding relaxation scheme on the fine grid only). It is found that MGM becomes significantly more effective than SOR as more accuracy is demanded and as more corrective grids, or more grid points, are used. For the accuracy required, it is found that MGM is two to three times faster than SOR in computing time. With the four-color relaxation scheme as applied to the impinging jet problem, the advantage of MGM over SOR is not as great. This may be due to the effect of a poor initial guess on MGM for this problem.

Ohring, S.↗

Bulk processing techniques for very large areas - Landsat classification of California

In 1977, California Law AB452 was passed to provide a mandate for the California Department of Forestry (CDF) to design and implement an information system to assess the forest land base for multiple uses and values. In connection with this mandate, a land-cover map of the entire state, emphasizing forest types, was produced. In producing this map, the latest techniques in digital image mosaicking were combined with the highspeed processing capability available on the ILLIAC IV parallel processor and other computer systems at the Ames Research Center (ARC). An operational and very responsive analysis method was developed at ARC that permitted on-time response to weekly workshops conducted with CDF field personnel to identify all 1,200 spectral classes and to produce final products. Over 100,000,000 acres were classified in the period between December 1, 1978, and April 15, 1979. All analyses were conducted using existing software.

Newland, W.↗

SAR processing on the MPP

The processing of synthetic aperture radar (SAR) signals using the massively parallel processor (MPP) is discussed. The fast Fourier transform convolution procedures employed in the algorithms are described. The MPP architecture comprises an array unit (ARU) which processes arrays of data; an array control unit which controls the operation of the ARU and performs scalar arithmetic; a program and data management unit which controls the flow of data; and a unique staging memory (SM) which buffers and permutes data. The ARU contains a 128 by 128 array of bit-serial processing elements (PE). Two-by-four surarrays of PE's are packaged in a custom VLSI HCMOS chip. The staging memory is a large multidimensional-access memory which buffers and permutes data flowing with the system. Efficient SAR processing is achieved via ARU communication paths and SM data manipulation. Real time processing capability can be realized via a multiple ARU, multiple SM configuration.

Batcher, K. E.↗

A linear decomposition method for large optimization problems. Blueprint for development

A method is proposed for decomposing large optimization problems encountered in the design of engineering systems such as an aircraft into a number of smaller subproblems. The decomposition is achieved by organizing the problem and the subordinated subproblems in a tree hierarchy and optimizing each subsystem separately. Coupling of the subproblems is accounted for by subsequent optimization of the entire system based on sensitivities of the suboptimization problem solutions at each level of the tree to variables of the next higher level. A formalization of the procedure suitable for computer implementation is developed and the state of readiness of the implementation building blocks is reviewed showing that the ingredients for the development are on the shelf. The decomposition method is also shown to be compatible with the natural human organization of the design process of engineering systems. The method is also examined with respect to the trends in computer hardware and software progress to point out that its efficiency can be amplified by network computing using parallel processors.

Sobieszczanski-Sobieski, J.↗

Advancements in real-time engine simulation technology

The approaches used to develop real-time engine simulations are reviewed. Both digital and hybrid (analog and digital) techniques are discussed and specific examples of each are cited. These approaches are assessed from the standpoint of their usefulness for digital engine control development. A number of NASA-sponsored simulation research activities, aimed at exploring real-time simulation techniques, are described. These include the development of a microcomputer-based, parallel processor system for real-time engine simulation.

Szuch, J. R.↗

Beyond the supercomputer

A NASA-directed development of massively parallel processor (MPP) computers is outlined, noting intended applications for data processing for near term earth resource and environment mapping, radar, and television transmissions. The MPP is designed to perform 100 billion operations/sec to obtain satisfactory image processing, while separate processing units correct distortions, register images, calculate correlation functions, and classify multispectral characteristics. Arrays of 1s and 0s will be manipulated in analog-to-digital conversions generating separate planes corresponding to powers of binaries. Data wires are replaced by fiber-optic tubes or thousands of wires, and single logic gates are replaced by thousands of logic gates and every memory element by thousands of memory elements. Features of the interconnections and the images control processor units are detailed, along with implementation of sliders for program flexibility.

Schaefer, D. H.↗

Interactive computer modeling of combustion chemistry and coalescence-dispersion modeling of turbulent combustion

An interactive computer code for simulation of a high-intensity turbulent combustor as a single point inhomogeneous stirred reactor was developed from an existing batch processing computer code CDPSR. The interactive CDPSR code was used as a guide for interpretation and direction of DOE-sponsored companion experiments utilizing Xenon tracer with optical laser diagnostic techniques to experimentally determine the appropriate mixing frequency, and for validation of CDPSR as a mixing-chemistry model for a laboratory jet-stirred reactor. The coalescence-dispersion model for finite rate mixing was incorporated into an existing interactive code AVCO-MARK I, to enable simulation of a combustor as a modular array of stirred flow and plug flow elements, each having a prescribed finite mixing frequency, or axial distribution of mixing frequency, as appropriate. Further increase the speed and reliability of the batch kinetics integrator code CREKID was increased by rewriting in vectorized form for execution on a vector or parallel processor, and by incorporating numerical techniques which enhance execution speed by permitting specification of a very low accuracy tolerance.

Pratt, D. T.↗