Search NASA⌕ Search

SEARCH · Search NASA

Results for “parallel computing algorithms”

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 757 records · Page 42

Introduction to a system for implementing neural net connections on SIMD architectures

Neural networks have attracted much interest recently, and using parallel architectures to simulate neural networks is a natural and necessary application. The SIMD model of parallel computation is chosen, because systems of this type can be built with large numbers of processing elements. However, such systems are not naturally suited to generalized communication. A method is proposed that allows an implementation of neural network connections on massively parallel SIMD architectures. The key to this system is an algorithm permitting the formation of arbitrary connections between the neurons. A feature is the ability to add new connections quickly. It also has error recovery ability and is robust over a variety of network topologies. Simulations of the general connection system, and its implementation on the Connection Machine, indicate that the time and space requirements are proportional to the product of the average number of connections per neuron and the diameter of the interconnection network.

Tomboulian, Sherryl↗

Introduction to a system for implementing neural net connections on SIMD architectures

Neural networks have attracted much interest recently, and using parallel architectures to simulate neural networks is a natural and necessary application. The SIMD model of parallel computation is chosen, because systems of this type can be built with large numbers of processing elements. However, such systems are not naturally suited to generalized elements. A method is proposed that allows an implementation of neural network connections on massively parallel SIMD architectures. The key to this system is an algorithm permitting the formation of arbitrary connections between the neurons. A feature is the ability to add new connections quickly. It also has error recovery ability and is robust over a variety of network topologies. Simulations of the general connection system, and its implementation on the Connection Machine, indicate that the time and space requirements are proportional to the product of the average number of connections per neuron and the diameter of the interconnection network.

Tomboulian, Sherryl↗

Execution time supports for adaptive scientific algorithms on distributed memory machines

Optimizations are considered that are required for efficient execution of code segments that consists of loops over distributed data structures. The PARTI (Parallel Automated Runtime Toolkit at ICASE) execution time primitives are designed to carry out these optimizations and can be used to implement a wide range of scientific algorithms on distributed memory machines. These primitives allow the user to control array mappings in a way that gives an appearance of shared memory. Computations can be based on a global index set. Primitives are used to carry out gather and scatter operations on distributed arrays. Communications patterns are derived at runtime, and the appropriate send and receive messages are automatically generated.

Berryman, Harry↗

Execution time support for scientific programs on distributed memory machines

Optimizations are considered that are required for efficient execution of code segments that consists of loops over distributed data structures. The PARTI (Parallel Automated Runtime Toolkit at ICASE) execution time primitives are designed to carry out these optimizations and can be used to implement a wide range of scientific algorithms on distributed memory machines. These primitives allow the user to control array mappings in a way that gives an appearance of shared memory. Computations can be based on a global index set. Primitives are used to carry out gather and scatter operations on distributed arrays. Communications patterns are derived at runtime, and the appropriate send and receive messages are automatically generated.

Berryman, Harry↗

Optimal processor assignment for pipeline computations

The availability of large scale multitasked parallel architectures introduces the following processor assignment problem for pipelined computations. Given a set of tasks and their precedence constraints, along with their experimentally determined individual responses times for different processor sizes, find an assignment of processor to tasks. Two objectives are of interest: minimal response given a throughput requirement, and maximal throughput given a response time requirement. These assignment problems differ considerably from the classical mapping problem in which several tasks share a processor; instead, it is assumed that a large number of processors are to be assigned to a relatively small number of tasks. Efficient assignment algorithms were developed for different classes of task structures. For a p processor system and a series parallel precedence graph with n constituent tasks, an O(np2) algorithm is provided that finds the optimal assignment for the response time optimization problem; it was found that the assignment optimizing the constrained throughput in O(np2log p) time. Special cases of linear, independent, and tree graphs are also considered.

Nicol, David M.↗

HTMT-class Latency Tolerant Parallel Architecture for Petaflops Scale Computation

Computational Aero Sciences and other numeric intensive computation disciplines demand computing throughputs substantially greater than the Teraflops scale systems only now becoming available. The related fields of fluids, structures, thermal, combustion, and dynamic controls are among the interdisciplinary areas that in combination with sufficient resolution and advanced adaptive techniques may force performance requirements towards Petaflops. This will be especially true for compute intensive models such as Navier-Stokes are or when such system models are only part of a larger design optimization computation involving many design points. Yet recent experience with conventional MPP configurations comprising commodity processing and memory components has shown that larger scale frequently results in higher programming difficulty and lower system efficiency. While important advances in system software and algorithms techniques have had some impact on efficiency and programmability for certain classes of problems, in general it is unlikely that software alone will resolve the challenges to higher scalability. As in the past, future generations of high-end computers may require a combination of hardware architecture and system software advances to enable efficient operation at a Petaflops level. The NASA led HTMT project has engaged the talents of a broad interdisciplinary team to develop a new strategy in high-end system architecture to deliver petaflops scale computing in the 2004/5 timeframe. The Hybrid-Technology, MultiThreaded parallel computer architecture incorporates several advanced technologies in combination with an innovative dynamic adaptive scheduling mechanism to provide unprecedented performance and efficiency within practical constraints of cost, complexity, and power consumption. The emerging superconductor Rapid Single Flux Quantum electronics can operate at 100 GHz (the record is 770 GHz) and one percent of the power required by convention semiconductor logic. Wave Division Multiplexing optical communications can approach a peak per fiber bandwidth of 1 Tbps and the new Data Vortex network topology employing this technology can connect tens of thousands of ports providing a bi-section bandwidth on the order of a Petabyte per second with latencies well below 100 nanoseconds, even under heavy loads. Processor-in-Memory (PIM) technology combines logic and memory on the same chip exposing the internal bandwidth of the memory row buffers at low latency. And holographic storage photorefractive storage technologies provide high-density memory with access a thousand times faster than conventional disk technologies. Together these technologies enable a new class of shared memory system architecture with a peak performance in the range of a Petaflops but size and power requirements comparable to today's largest Teraflops scale systems. To achieve high-sustained performance, HTMT combines an advanced multithreading processor architecture with a memory-driven coarse-grained latency management strategy called "percolation", yielding high efficiency while reducing the much of the parallel programming burden. This paper will present the basic system architecture characteristics made possible through this series of advanced technologies and then give a detailed description of the new percolation approach to runtime latency management.

Sterling, Thomas↗

Numerically stable algorithm for discrete-ordinate-method radiative transfer in multiple scattering and emitting layered media

The transfer of monochromatic radiation in a scattering, absorbing, and emitting plane-parallel medium with a specified bidirectional reflectivity at the lower boundary is considered. The equations and boundary conditions are summarized. The numerical implementation of the theory is discussed with attention given to the reliable and efficient computation of eigenvalues and eigenvectors. Ways of avoiding fatal overflows and ill-conditioning in the matrix inversion needed to determine the integration constants are also presented.

Stamnes, Knut↗

Neural network for processing both spatial and temporal data with time based back-propagation

Neural networks are computing systems modeled after the paradigm of the biological brain. For years, researchers using various forms of neural networks have attempted to model the brain's information processing and decision-making capabilities. Neural network algorithms have impressively demonstrated the capability of modeling spatial information. On the other hand, the application of parallel distributed models to the processing of temporal data has been severely restricted. The invention introduces a novel technique which adds the dimension of time to the well known back-propagation neural network algorithm. In the space-time neural network disclosed herein, the synaptic weights between two artificial neurons (processing elements) are replaced with an adaptable-adjustable filter. Instead of a single synaptic weight, the invention provides a plurality of weights representing not only association, but also temporal dependencies. In this case, the synaptic weights are the coefficients to the adaptable digital filters. Novelty is believed to lie in the disclosure of a processing element and a network of the processing elements which are capable of processing temporal as well as spacial data.

Villarreal, James A.↗

Comparison of ERBE inferred and model computed clear-sky albedos

Over-ocean clear-sky albedos measured with instruments on the Earth Radiation Budget Satellite (ERBS) are compared with albedos simulated using a radiative transfer model (RTM). The comparison covers the monthly mean albedos for November 1984. The ERBS albedo was calculated with a scene identification algorithm. Techniques used to suppress cloud cover uncertainties are discussed. The plane-parallel delta-Eddington RTM accounted for O3, O2, CO2 and H2O gaseous absorption and background aerosol absorption.

Briegleb, B. P.↗

Slices: A Scalable Partitioner for Finite Element Meshes

A parallel partitioner for partitioning unstructured finite element meshes on distributed memory architectures is developed. The element based partitioner can handle mixtures of different element types. All algorithms adopted in the partitioner are scalable, including a communication template for unpredictable incoming messages, as shown in actual timing measurements.

partitioner finite element meshes parallel computi↗

Radiative transfer calculated from a Markov chain formalism

The theory of Markov chains is used to formulate the radiative transport problem in a general way by modeling the successive interactions of a photon as a stochastic process. Under the minimal requirement that the stochastic process is a Markov chain, the determination of the diffuse reflection or transmission from a scattering atmosphere is equivalent to the solution of a system of linear equations. This treatment is mathematically equivalent to, and thus has many of the advantages of, Monte Carlo methods, but can be considerably more rapid than Monte Carlo algorithms for numerical calculations in particular applications. We have verified the speed and accuracy of this formalism for the standard problem of finding the intensity of scattered light from a homogeneous plane-parallel atmosphere with an arbitrary phase function for scattering. Accurate results over a wide range of parameters were obtained with computation times comparable to those of a standard 'doubling' routine. The generality of this formalism thus allows fast, direct solutions to problems that were previously soluble only by Monte Carlo methods. Some comparisons are made with respect to integral equation methods.

Esposito, L. W.↗

Evolutionary Computational Methods for Identifying Emergent Behavior in Autonomous Systems

A technique based on Evolutionary Computational Methods (ECMs) was developed that allows for the automated optimization of complex computationally modeled systems, such as autonomous systems. The primary technology, which enables the ECM to find optimal solutions in complex search spaces, derives from evolutionary algorithms such as the genetic algorithm and differential evolution. These methods are based on biological processes, particularly genetics, and define an iterative process that evolves parameter sets into an optimum. Evolutionary computation is a method that operates on a population of existing computational-based engineering models (or simulators) and competes them using biologically inspired genetic operators on large parallel cluster computers. The result is the ability to automatically find design optimizations and trades, and thereby greatly amplify the role of the system engineer.

Terrile, Richard J.↗

A Lagrange multiplier based divide and conquer finite element algorithm

A novel domain decomposition method based on a hybrid variational principle is presented. Prior to any computation, a given finite element mesh is torn into a set of totally disconnected submeshes. First, an incomplete solution is computed in each subdomain. Next, the compatibility of the displacement field at the interface nodes is enforced via discrete, polynomial and/or piecewise polynomial Lagrange multipliers. In the static case, each floating subdomain induces a local singularity that is resolved very efficiently. The interface problem associated with this domain decomposition method is, in general, indefinite and of variable size. A dedicated conjugate projected gradient algorithm is developed for solving the latter problem when it is not feasible to explicitly assemble the interface operator. When implemented on local memory multiprocessors, the proposed methodology requires less interprocessor communication than the classical method of substructuring. It is also suitable for parallel/vector computers with shared memory and compares favorably with factorization based parallel direct methods.

Farhat, C.↗

The star identification, pointing and tracking system of UVSTAR, an attached payload instrument system for the Shuttle Hitchhiker-M platform

We describe an algorithm for star identification and pointing/tracking of a spaceborne electro-optical system and simulation analyses to test the algorithm. The algorithm will be implemented in the guiding system of UVSTAR, a spectrographic telescope for observations of astronomical and planetary sources operating in the 500-1250 A waveband at approximately 1 A resolution. The experiment is an attached payload and will fly as a Hitchhiker-M payload on the Shuttle. UVSTAR includes capabilities for independent target acquisition and tracking. The spectrograph package has internal gimbals that allow angular movement of plus or minus 3 deg from the central position. Rotation about the azimuth axis (parallel to the Shuttle z axis) and elevation axis (parallel to the Shuttle x axis) will actively position the field of view to center the target of interest in the fields of the spectrographs. The algorithm is based on an on-board catalog of stars. To identify star fields, the algorithm compares the positions of stars recorded by the guiding imager to positions computed from the on-board catalog. When the field has been identified, its position within the guiding imager field of view can be used to compute the pointing corrections necessary to point to a target of interest. In tracking mode, the software uses the past history to predict the quasi-periodic attitude control motions of the shuttle and sends pointing commands to cancel the motion and stabilize UVSTAR on the target. The guiding imager (guider) will have an 80-mm focal length and f/1.4 optics giving a field of view of 6 deg x 4.5 deg using a 385 x 288 pixel intensified CCD. It will be capable of providing high accuracy (better than 2 arc-sec) attitude determination from coarse (6 deg x 4.5 deg) initial knowledge of the pointing direction; and of pointing toward the target. It will also be capable of tracking at the same high accuracy with a processing time of less than a few hundredths of a second.

Decarlo, Francesco↗

Adding dynamic rules to self-organizing fuzzy systems

This paper develops a Dynamic Self-Organizing Fuzzy System (DSOFS) capable of adding, removing, and/or adapting the fuzzy rules and the fuzzy reference sets. The DSOFS background consists of a self-organizing neural structure with neuron relocation features which will develop a map of the input-output behavior. The relocation algorithm extends the topological ordering concept. Fuzzy rules (neurons) are dynamically added or released while the neural structure learns the pattern. The DSOFS advantages are the automatic synthesis and the possibility of parallel implementation. A high adaptation speed and a reduced number of neurons is needed in order to keep errors under some limits. The computer simulation results are presented in a nonlinear systems modelling application.

Buhusi, Catalin V.↗

Route planning in a four-dimensional environment

Robots must be able to function in the real world. The real world involves processes and agents that move independently of the actions of the robot, sometimes in an unpredictable manner. A real-time integrated route planning and spatial representation system for planning routes through dynamic domains is presented. The system will find the safest most efficient route through space-time as described by a set of user defined evaluation functions. Because the route planning algorthims is highly parallel and can run on an SIMD machine in O(p) time (p is the length of a path), the system will find real-time paths through unpredictable domains when used in an incremental mode. Spatial representation, an SIMD algorithm for route planning in a dynamic domain, and results from an implementation on a traditional computer architecture are discussed.

Slack, M. G.↗

NETRA: A parallel architecture for integrated vision systems 2: Algorithms and performance evaluation

In part 1 architecture of NETRA is presented. A performance evaluation of NETRA using several common vision algorithms is also presented. Performance of algorithms when they are mapped on one cluster is described. It is shown that SIMD, MIMD, and systolic algorithms can be easily mapped onto processor clusters, and almost linear speedups are possible. For some algorithms, analytical performance results are compared with implementation performance results. It is observed that the analysis is very accurate. Performance analysis of parallel algorithms when mapped across clusters is presented. Mappings across clusters illustrate the importance and use of shared as well as distributed memory in achieving high performance. The parameters for evaluation are derived from the characteristics of the parallel algorithms, and these parameters are used to evaluate the alternative communication strategies in NETRA. Furthermore, the effect of communication interference from other processors in the system on the execution of an algorithm is studied. Using the analysis, performance of many algorithms with different characteristics is presented. It is observed that if communication speeds are matched with the computation speeds, good speedups are possible when algorithms are mapped across clusters.

Choudhary, Alok N.↗

Bit-parallel arithmetic in a massively-parallel associative processor

A simple but powerful new architecture based on a classical associative processor model is presented. Algorithms for performing the four basic arithmetic operations both for integer and floating point operands are described. For m-bit operands, the proposed architecture makes it possible to execute complex operations in O(m) cycles as opposed to O(m exp 2) for bit-serial machines. A word-parallel, bit-parallel, massively-parallel computing system can be constructed using this architecture with VLSI technology. The operation of this system is demonstrated for the fast Fourier transform and matrix multiplication.

Scherson, Isaac D.↗