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 559 records · Page 31

Efficient Multidisciplinary Analysis Procedure Using Multi-Level Parallelization Approach

Multidisciplinary applications are suitable for parallel computing environment by adopting the domain decomposition method. Immediately, a multidisciplinary application can be parallelized by solving each discipline separately. In order to perform coupled multidisciplinary analysis, coupling of each discipline can be accomplished by exchanging boundary data at the interfaces. This is regarded as discipline-level parallelization. Next level could be a "coarse-grain" parallelization of each discipline, which mainly depends on the physical geometry and nature of each discipline. For example, it is almost impossible for structured-grid based computational fluid dynamics codes to do flow analysis of an aircraft by using a single grid because of the complexity of its configuration. Thus, multi-block grid is commonly used to describe the details of complex geometry. Similarly, in structural analysis, the structure is frequently subdivided into substructures. Thus, the computation of each subdomain can be easily parallelized since each subdomain is solved separately independent of other domains. The parallelization is accomplished by solving each subdomain separately on a separate processor and exchanging the boundary conditions at domain interfaces periodically. However, the physical decomposition of the domain introduces explicit boundary conditions at the domain interfaces. This is not desirable for critical areas such as those containing shock waves or flow separations. Thus, a "fine-grain" parallelization is introduced to overcome this problem. The "fine-grain" parallelization is one that solves exactly the same system of equations of a subdomain by using more than one processors without introducing any explicit boundary conditions. An efficient multidisciplinary analysis procedure can be accomplished by successfully combining the above multi-level parallelism. A multidisciplinary analysis code, ENSAERO developed at NASA Ames Research Center is used in this study to implement the proposed approach. The communication data structure required for the proposed approach will be studied in detail. This work will demonstrate the feasibility of using multi-level parallelization approach in multidisciplinary analysis applications.

Byun, Chansup↗

Comparison of Origin 2000 and Origin 3000 Using NAS Parallel Benchmarks

This report describes results of benchmark tests on the Origin 3000 system currently being installed at the NASA Ames National Advanced Supercomputing facility. This machine will ultimately contain 1024 R14K processors. The first part of the system, installed in November, 2000 and named mendel, is an Origin 3000 with 128 R12K processors. For comparison purposes, the tests were also run on lomax, an Origin 2000 with R12K processors. The BT, LU, and SP application benchmarks in the NAS Parallel Benchmark Suite and the kernel benchmark FT were chosen to determine system performance and measure the impact of changes on the machine as it evolves. Having been written to measure performance on Computational Fluid Dynamics applications, these benchmarks are assumed appropriate to represent the NAS workload. Since the NAS runs both message passing (MPI) and shared-memory, compiler directive type codes, both MPI and OpenMP versions of the benchmarks were used. The MPI versions used were the latest official release of the NAS Parallel Benchmarks, version 2.3. The OpenMP versiqns used were PBN3b2, a beta version that is in the process of being released. NPB 2.3 and PBN 3b2 are technically different benchmarks, and NPB results are not directly comparable to PBN results.

Turney, Raymond D.↗

Opto-electronic morphological processor

The opto-electronic morphological processor of the present invention is capable of receiving optical inputs and emitting optical outputs. The use of optics allows implementation of parallel input/output, thereby overcoming a major bottleneck in prior art image processing systems. The processor consists of three components, namely, detectors, morphological operators and modulators. The detectors and operators are fabricated on a silicon VLSI chip and implement the optical input and morphological operations. A layer of ferro-electric liquid crystals is integrated with a silicon chip to provide the optical modulation. The implementation of the image processing operators in electronics leads to a wide range of applications and the use of optical connections allows cascadability of these parallel opto-electronic image processing components and high speed operation. Such an opto-electronic morphological processor may be used as the pre-processing stage in an image recognition system. In one example disclosed herein, the optical input/optical output morphological processor of the invention is interfaced with a binary phase-only correlator to produce an image recognition system.

Yu, Jeffrey W.↗

LU and Cholesky decomposition on an optical systolic array processor

Direct solutions of matrix-vector equations on an optical systolic array processor are considered. The solutions are discussed and a parallel algorithm for LU matrix decomposition that is very attractive for an optical realization is formulated. It is noted that when direct techniques are used, it is preferable to realize the matrix decomposition on an optical system and to utilize a digital processor for the solution of the simplified resultant matrix-vector problem. One method of realizing LU matrix decomposition on a new frequency-multiplexed optical systolic array matrix-matrix processor is described. A simple method for extending the process of LU decomposition to Cholesky decomposition on the optical processor is discussed.

Casasent, D.↗

Three-dimensional radiative transfer on a massively parallel computer

We perform 3D radiative transfer calculations in non-local thermodynamic equilibrium (NLTE) in the simple two-level atom approximation on the Mas-Par MP-1, which contains 8192 processors and is a single instruction multiple data (SIMD) machine, an example of the new generation of massively parallel computers. On such a machine, all processors execute the same command at a given time, but on different data. To make radiative transfer calculations efficient, we must re-consider the numerical methods and storage of data. To solve the transfer equation, we adopt the short characteristic method and examine different acceleration methods to obtain the source function. We use the ALI method and test local and non-local operators. Furthermore, we compare the Ng and the orthomin methods of acceleration. We also investigate the use of multi-grid methods to get fast solutions for the NLTE case. In order to test these numerical methods, we apply them to two problems with and without periodic boundary conditions.

Vath, H. M.↗

On the impact of communication complexity in the design of parallel numerical algorithms

This paper describes two models of the cost of data movement in parallel numerical algorithms. One model is a generalization of an approach due to Hockney, and is suitable for shared memory multiprocessors where each processor has vector capabilities. The other model is applicable to highly parallel nonshared memory MIMD systems. In the second model, algorithm performance is characterized in terms of the communication network design. Techniques used in VLSI complexity theory are also brought in, and algorithm independent upper bounds on system performance are derived for several problems that are important to scientific computation.

Gannon, D.↗

On the impact of communication complexity on the design of parallel numerical algorithms

This paper describes two models of the cost of data movement in parallel numerical alorithms. One model is a generalization of an approach due to Hockney, and is suitable for shared memory multiprocessors where each processor has vector capabilities. The other model is applicable to highly parallel nonshared memory MIMD systems. In this second model, algorithm performance is characterized in terms of the communication network design. Techniques used in VLSI complexity theory are also brought in, and algorithm-independent upper bounds on system performance are derived for several problems that are important to scientific computation.

Gannon, D. B.↗

Parallel simulated annealing algorithms for cell placement on hypercube multiprocessors

Two parallel algorithms for standard cell placement using simulated annealing are developed to run on distributed-memory message-passing hypercube multiprocessors. The cells can be mapped in a two-dimensional area of a chip onto processors in an n-dimensional hypercube in two ways, such that both small and large cell exchange and displacement moves can be applied. The computation of the cost function in parallel among all the processors in the hypercube is described, along with a distributed data structure that needs to be stored in the hypercube to support the parallel cost evaluation. A novel tree broadcasting strategy is used extensively for updating cell locations in the parallel environment. A dynamic parallel annealing schedule estimates the errors due to interacting parallel moves and adapts the rate of synchronization automatically. Two novel approaches in controlling error in parallel algorithms are described: heuristic cell coloring and adaptive sequence control.

Banerjee, Prithviraj↗

Modula-2*: An extension of Modula-2 for highly parallel programs

Parallel programs should be machine-independent, i.e., independent of properties that are likely to differ from one parallel computer to the next. Extensions are described of Modula-2 for writing highly parallel, portable programs meeting these requirements. The extensions are: synchronous and asynchronous forms of forall statement; and control of the allocation of data to processors. Sample programs written with the extensions demonstrate the clarity of parallel programs when machine-dependent details are omitted. The principles of efficiently implementing the extensions on SIMD, MIMD, and MSIMD machines are discussed. The extensions are small enough to be integrated easily into other imperative languages.

Tichy, Walter F.↗

Parallel computation of geometry control in adaptive truss structures

The fast computation of geometry control in adaptive truss structures involves two distinct parts: the efficient integration of the inverse kinematic differential equations that govern the geometry control and the fast computation of the Jacobian, which appears on the right-hand-side of the inverse kinematic equations. This paper present an efficient parallel implementation of the Jacobian computation on an MIMD machine. Large speedup from the parallel implementation is obtained, which reduces the Jacobian computation to an O(M-squared/n) procedure on an n-processor machine, where M is the number of members in the adaptive truss. The parallel algorithm given here is a good candidate for on-line geometry control of adaptive structures using attached processors.

Ramesh, A. V.↗

A New Approach to Parallel Dynamic Partitioning for Adaptive Unstructured Meshes

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.

Heber, Gerd↗

An implementation of a tree code on a SIMD, parallel computer

We describe a fast tree algorithm for gravitational N-body simulation on SIMD parallel computers. The tree construction uses fast, parallel sorts. The sorted lists are recursively divided along their x, y and z coordinates. This data structure is a completely balanced tree (i.e., each particle is paired with exactly one other particle) and maintains good spatial locality. An implementation of this tree-building algorithm on a 16k processor Maspar MP-1 performs well and constitutes only a small fraction (approximately 15%) of the entire cycle of finding the accelerations. Each node in the tree is treated as a monopole. The tree search and the summation of accelerations also perform well. During the tree search, node data that is needed from another processor is simply fetched. Roughly 55% of the tree search time is spent in communications between processors. We apply the code to two problems of astrophysical interest. The first is a simulation of the close passage of two gravitationally, interacting, disk galaxies using 65,636 particles. We also simulate the formation of structure in an expanding, model universe using 1,048,576 particles. Our code attains speeds comparable to one head of a Cray Y-MP, so single instruction, multiple data (SIMD) type computers can be used for these simulations. The cost/performance ratio for SIMD machines like the Maspar MP-1 make them an extremely attractive alternative to either vector processors or large multiple instruction, multiple data (MIMD) type parallel computers. With further optimizations (e.g., more careful load balancing), speeds in excess of today's vector processing computers should be possible.

Olson, Kevin M.↗

A parallel simulated annealing algorithm for standard cell placement on a hypercube computer

A parallel version of a simulated annealing algorithm is presented which is targeted to run on a hypercube computer. A strategy for mapping the cells in a two dimensional area of a chip onto processors in an n-dimensional hypercube is proposed such that both small and large distance moves can be applied. Two types of moves are allowed: cell exchanges and cell displacements. The computation of the cost function in parallel among all the processors in the hypercube is described along with a distributed data structure that needs to be stored in the hypercube to support parallel cost evaluation. A novel tree broadcasting strategy is used extensively in the algorithm for updating cell locations in the parallel environment. Studies on the performance of the algorithm on example industrial circuits show that it is faster and gives better final placement results than the uniprocessor simulated annealing algorithms. An improved uniprocessor algorithm is proposed which is based on the improved results obtained from parallelization of the simulated annealing algorithm.

Jones, Mark Howard↗

Parallel matrix multiplication on the Connection Machine

Matrix multiplication is a computation and communication intensive problem. Six parallel algorithms for matrix multiplication on the Connection Machine are presented and compared with respect to their performance and processor usage. For n by n matrices, the algorithms have theoretical running times of O(n to the 2nd power log n), O(n log n), O(n), and O(log n), and require n, n to the 2nd power, n to the 2nd power, and n to the 3rd power processors, respectively. With careful attention to communication patterns, the theoretically predicted runtimes can indeed be achieved in practice. The parallel algorithms illustrate the tradeoffs between performance, communication cost, and processor usage.

Tichy, Walter F.↗

Parallel Domain Decomposition Formulation and Software for Large-Scale Sparse Symmetrical/Unsymmetrical Aeroacoustic Applications

The overall objectives of this research work are to formulate and validate efficient parallel algorithms, and to efficiently design/implement computer software for solving large-scale acoustic problems, arised from the unified frameworks of the finite element procedures. The adopted parallel Finite Element (FE) Domain Decomposition (DD) procedures should fully take advantages of multiple processing capabilities offered by most modern high performance computing platforms for efficient parallel computation. To achieve this objective. the formulation needs to integrate efficient sparse (and dense) assembly techniques, hybrid (or mixed) direct and iterative equation solvers, proper pre-conditioned strategies, unrolling strategies, and effective processors' communicating schemes. Finally, the numerical performance of the developed parallel finite element procedures will be evaluated by solving series of structural, and acoustic (symmetrical and un-symmetrical) problems (in different computing platforms). Comparisons with existing "commercialized" and/or "public domain" software are also included, whenever possible.

Nguyen, D. T.↗

Parallel-Access Alinement Network Using Barrel Switches

Practical version of parallel-access alinement network utilizes two barrel switches for interfacing N parallel memory modules with N parallel processing elements. Switches are interconnected where 17 memory ports (MP's) are connected to 17 processor ports (PP's). Network uses two electronic barrel switches to direct data flow in parallel data-processing system. Each switch can shift a multibit parallel input a predetermined number of places to left or right, end off, or end around in one clock-pulse.

Barnes, George H.↗

Performance Analysis and Portability of the PLUM Load Balancing System

The ability to dynamically adapt an unstructured mesh is a powerful tool for solving computational problems with evolving physical features; however, an efficient parallel implementation is rather difficult. To address this problem, we have developed PLUM, an automatic portable framework for performing adaptive numerical computations in a message-passing environment. PLUM requires that all data be globally redistributed after each mesh adaption to achieve load balance. We present an algorithm for minimizing this remapping overhead by guaranteeing an optimal processor reassignment. We also show that the data redistribution cost can be significantly reduced by applying our heuristic processor reassignment algorithm to the default mapping of the parallel partitioner. Portability is examined by comparing performance on a SP2, an Origin2000, and a T3E. Results show that PLUM can be successfully ported to different platforms without any code modifications.

Oliker, Leonid↗

Parallel algorithms and architecture for computation of manipulator forward dynamics

Parallel computation of manipulator forward dynamics is investigated. Considering three classes of algorithms for the solution of the problem, that is, the O(n), the O(n exp 2), and the O(n exp 3) algorithms, parallelism in the problem is analyzed. It is shown that the problem belongs to the class of NC and that the time and processors bounds are of O(log2/2n) and O(n exp 4), respectively. However, the fastest stable parallel algorithms achieve the computation time of O(n) and can be derived by parallelization of the O(n exp 3) serial algorithms. Parallel computation of the O(n exp 3) algorithms requires the development of parallel algorithms for a set of fundamentally different problems, that is, the Newton-Euler formulation, the computation of the inertia matrix, decomposition of the symmetric, positive definite matrix, and the solution of triangular systems. Parallel algorithms for this set of problems are developed which can be efficiently implemented on a unique architecture, a triangular array of n(n+2)/2 processors with a simple nearest-neighbor interconnection. This architecture is particularly suitable for VLSI and WSI implementations. The developed parallel algorithm, compared to the best serial O(n) algorithm, achieves an asymptotic speedup of more than two orders-of-magnitude in the computation the forward dynamics.

Fijany, Amir↗