The design of a highly parallel computer organization.
Computer organization design employing distributed processor to provide general purpose computing while taking advantage of large scale integration
Koczela, L. J.↗
SEARCH · Search NASA
Results for “parallel computing”
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.
Parallel computation of unsteady, three-dimensional, chemically reacting, nonequilibrium flow using a time-split finite-volume method on the Illiac IV
A description is presented of the split finite-volume method which is a viable numerical procedure for performing with the aid of a modern special purpose vector computer numerical simulation studies of complicated flow fields, including chemical reactions, about geometrically complex bodies. Such numerical studies are needed for the development of atmospheric entry vehicles such as the space shuttle. The equations which are approximated are quite general and can be used in studies of combustion, pollution, and other chemically reacting flow phenomena, where convective transport effects dominate the influence of radiative, viscous, and other transport mechanisms. The shock perturbed flow about a shuttle orbiter flying at a large angle of attack during atmospheric entry is illustrated. The method uses a time splitting of the convection differencing operator to achieve efficient data management.
Reinhardt, W. A.↗
Dynamic remapping of parallel computations with varying resource demands
A large class of computational problems is characterized by frequent synchronization, and computational requirements which change as a function of time. When such a problem must be solved on a message passing multiprocessor machine, the combination of these characteristics lead to system performance which decreases in time. Performance can be improved with periodic redistribution of computational load; however, redistribution can exact a sometimes large delay cost. We study the issue of deciding when to invoke a global load remapping mechanism. Such a decision policy must effectively weigh the costs of remapping against the performance benefits. We treat this problem by constructing two analytic models which exhibit stochastically decreasing performance. One model is quite tractable; we are able to describe the optimal remapping algorithm, and the optimal decision policy governing when to invoke that algorithm. However, computational complexity prohibits the use of the optimal remapping decision policy. We then study the performance of a general remapping policy on both analytic models. This policy attempts to minimize a statistic W(n) which measures the system degradation (including the cost of remapping) per computation step over a period of n steps. We show that as a function of time, the expected value of W(n) has at most one minimum, and that when this minimum exists it defines the optimal fixed-interval remapping policy. Our decision policy appeals to this result by remapping when it estimates that W(n) is minimized. Our performance data suggests that this policy effectively finds the natural frequency of remapping. We also use the analytic models to express the relationship between performance and remapping cost, number of processors, and the computation's stochastic activity.
Nicol, D. M.↗
Time-partitioning simulation models for calculation on parallel computers
A technique allowing time-staggered solution of partial differential equations is presented in this report. Using this technique, called time-partitioning, simulation execution speedup is proportional to the number of processors used because all processors operate simultaneously, with each updating of the solution grid at a different time point. The technique is limited by neither the number of processors available nor by the dimension of the solution grid. Time-partitioning was used to obtain the flow pattern through a cascade of airfoils, modeled by the Euler partial differential equations. An execution speedup factor of 1.77 was achieved using a two processor Cray X-MP/24 computer.
Milner, Edward J.↗
Implementation of an ADI method on parallel computers
The implementation of an ADI method for solving the diffusion equation on three parallel/vector computers is discussed. The computers were chosen so as to encompass a variety of architectures. They are: the MPP, an SIMD machine with 16K bit serial processors; FLEX/32, an MIMD machine with 20 processors; and CRAY/2, an MIMD machine with four vector processors. The Gaussian elimination algorithm is used to solve a set of tridiagonal systems on the FLEX/32 and CRAY/2 while the cyclic elimination algorithm is used to solve these systems on the MPP. The implementation of the method is discussed in relation to these architectures and measures of the performance on each machine are given. Simple performance models are used to describe the performance. These models highlight the bottlenecks and limiting factors for this algorithm on these architectures. Finally, conclusions are presented.
Fatoohi, Raad A.↗
Implementation of an ADI method on parallel computers
In this paper the implementation of an ADI method for solving the diffusion equation on three parallel/vector computers is discussed. The computers were chosen so as to encompass a variety of architectures. They are the MPP, an SIMD machine with 16-Kbit serial processors; Flex/32, an MIMD machine with 20 processors; and Cray/2, an MIMD machine with four vector processors. The Gaussian elimination algorithm is used to solve a set of tridiagonal systems on the Flex/32 and Cray/2 while the cyclic elimination algorithm is used to solve these systems on the MPP. The implementation of the method is discussed in relation to these architectures and measures of the performance on each machine are given. Simple performance models are used to describe the performance. These models highlight the bottlenecks and limiting factors for this algorithm on these architectures. Finally conclusions are presented.
Fatoohi, Raad A.↗
Solution of structural analysis problems on a parallel computer
The problems of a blade-stiffened panel with a hole subjected to compression, and a deployable space mast subjected to tip loads, are treated through the application of FEM to model generation followed by the solution of a linear system of equations. Direct and iterative approaches to the solution of the linear systems are solved in turn; for the panel problems using varying numbers of processors, the incomplete Cholesky-conjugate gradient method was the fastest iterative method on all but two instances in which the number of processors was large.
Storaasli, Olaf↗
Time-partitioning simulation models for calculation of parallel computers
A technique allowing time-staggered solution of partial differential equations is presented in this report. Using this technique, called time-partitioning, simulation execution speedup is proportional to the number of processors used because all processors operate simultaneously, with each updating of the solution grid at a different time point. The technique is limited by neither the number of processors available nor by the dimension of the solution grid. Time-partitioning was used to obtain the flow pattern through a cascade of airfoils, modeled by the Euler partial differential equations. An execution speedup factor of 1.77 was achieved using a two processor Cray X-MP/24 computer.
Milner, Edward J.↗
Dynamic remapping of parallel computations with varying resource demands
The issue of deciding when to invoke a global load remapping mechanism is studied. Such a decision policy must effectively weigh the costs of remapping against the performance benefits, and should be general enough to apply automatically to a wide range of computations. The authors propose a general mapping decision heuristic, then study its effectiveness and its anticipated behavior on two very different models of load evolution. Assuming only that the remapping cost is known, this policy dynamically minimizes system degradation (including the cost of remapping) for each computation step. This policy is quite simple, choosing to remap when the first local minimum in the degradation function is detected. Simulations show that the decision obtained provides significantly better performance than that achieved by never remapping. The authors also observe that the average intermapping frequency is quite close to the optimal fixed remapping frequency.
Nicol, David M.↗
On evaluating parallel computer systems
A workshop was held in an attempt to program real problems on the MIT Static Data Flow Machine. Most of the architecture of the machine was specified but some parts were incomplete. The main purpose for the workshop was to explore principles for the evaluation of computer systems employing new architectures. Principles explored were: (1) evaluation must be an integral, ongoing part of a project to develop a computer of radically new architecture; (2) the evaluation should seek to measure the usability of the system as well as its performance; (3) users from the application domains must be an integral part of the evaluation process; and (4) evaluation results should be fed back into the design process. It is concluded that the general organizational principles are achievable in practice from this workshop.
Adams, George B., III↗
Explicit and implicit solution of the Navier-Stokes equations on a massively parallel computer
The design, implementation, and performance of a two-dimensional time-accurate Navier-Stokes solver for the CM2 supercomputer are described. The program uses a single processor for each grid point. Two different time-stepping methods have so far been implemented: an explicit third-order Runge-Kutta method and an implicit approximation-factorization method. The CM2 results are checked against those of a mature well-vectorized Cray 2 program, both for correctness and performance. The code is found to be correct, and the performance in some cases is up to several times that of the Cray 2.
Levit, Creon↗
Three parallel computation methods for structural vibration analysis
Explore the source record for details and available documents.
Storaasli, Olaf↗
Highly parallel computer architecture for robotic computation
In a computer having a large number of single instruction multiple data (SIMD) processors, each of the SIMD processors has two sets of three individual processor elements controlled by a master control unit and interconnected among a plurality of register file units where data is stored. The register files input and output data in synchronism with a minor cycle clock under control of two slave control units controlling the register file units connected to respective ones of the two sets of processor elements. Depending upon which ones of the register file units are enabled to store or transmit data during a particular minor clock cycle, the processor elements within an SIMD processor are connected in rings or in pipeline arrays, and may exchange data with the internal bus or with neighboring SIMD processors through interface units controlled by respective ones of the two slave control units.
Fijany, Amir↗
Domain decomposition: A bridge between nature and parallel computers
Domain decomposition is an intuitive organizing principle for a partial differential equation (PDE) computation, both physically and architecturally. However, its significance extends beyond the readily apparent issues of geometry and discretization, on one hand, and of modular software and distributed hardware, on the other. Engineering and computer science aspects are bridged by an old but recently enriched mathematical theory that offers the subject not only unity, but also tools for analysis and generalization. Domain decomposition induces function-space and operator decompositions with valuable properties. Function-space bases and operator splittings that are not derived from domain decompositions generally lack one or more of these properties. The evolution of domain decomposition methods for elliptically dominated problems has linked two major algorithmic developments of the last 15 years: multilevel and Krylov methods. Domain decomposition methods may be considered descendants of both classes with an inheritance from each: they are nearly optimal and at the same time efficiently parallelizable. Many computationally driven application areas are ripe for these developments. A progression is made from a mathematically informal motivation for domain decomposition methods to a specific focus on fluid dynamics applications. To be introductory rather than comprehensive, simple examples are provided while convergence proofs and algorithmic details are left to the original references; however, an attempt is made to convey their most salient features, especially where this leads to algorithmic insight.
Keyes, David E.↗
MAX - An advanced parallel computer for space applications
MAX is a fault-tolerant multicomputer hardware and software architecture designed to meet the needs of NASA spacecraft systems. It consists of conventional computing modules (computers) connected via a dual network topology. One network is used to transfer data among the computers and between computers and I/O devices. This network's topology is arbitrary. The second network operates as a broadcast medium for operating system synchronization messages and supports the operating system's Byzantine resilience. A fully distributed operating system supports multitasking in an asynchronous event and data driven environment. A large grain dataflow paradigm is used to coordinate the multitasking and provide easy control of concurrency. It is the basis of the system's fault tolerance and allows both static and dynamical location of tasks. Redundant execution of tasks with software voting of results may be specified for critical tasks. The dataflow paradigm also supports simplified software design, test and maintenance. A unique feature is a method for reliably patching code in an executing dataflow application.
Lewis, Blair F.↗
Non-parametric analysis of LANDSAT maps using neural nets and parallel computers
Nearest neighbor approaches and a new neural network, the Binary Diamond, are used for the classification of images of ground pixels obtained by LANDSAT satellite. The performances are evaluated by comparing classifications of a scene in the vicinity of Washington DC. The problem of optimal selection of categories is addressed as a step in the classification process.
Salu, Yehuda↗
Parallel computers - Estimate errors caused by imprecise data
A new approach to the problem of estimating errors caused by imprecise data is proposed in the context of software engineering. A software device is used to produce an ideal solution to the problem, when the computer is capable of computing errors of arbitrary programs. The software engineering aspect of this problem is to describe a device for computing the error estimates in software terms and then to provide precise numbers with error estimates to the user. The feasibility of the program capable of computing both some quantity and its error estimate in the range of possible measurement errors is demonstrated.
Kreinovich, Vladik↗
Time accurate application of the MacCormack 2-4 scheme on massively parallel computers
Many recent computational efforts in turbulence and acoustics research have used higher order numerical algorithms. One popular method has been the explicit MacCormack 2-4 scheme. The MacCormack 2-4 scheme is second order accurate in time and fourth order accurate in space, and is stable for CFL's below 2/3. Current research has shown that the method can give accurate results but does exhibit significant Gibbs phenomena at sharp discontinuities. The impact of adding Jameson type second, third, and fourth order artificial viscosity was examined here. Category 2 problems, the nonlinear traveling wave and the Riemann problem, were computed using a CFL number of 0.25. This research has found that dispersion errors can be significantly reduced or nearly eliminated by using a combination of second and third order terms in the damping. Use of second and fourth order terms reduced the magnitude of dispersion errors but not as effectively as the second and third order combination. The program was coded using Thinking Machine's CM Fortran, a variant of Fortran 90/High Performance Fortran, and was executed on a 2K CM-200. Simple extrapolation boundary conditions were used for both problems.
Hudson, Dale A.↗