Search NASA⌕ Search

SEARCH · Search NASA

Results for “VECTOR PROCESSING”

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 37 records · Page 2

A vector scanning processing technique for pulsed laser velocimetry

Pulsed laser sheet velocimetry yields nonintrusive measurements of two-dimensional velocity vectors across an extended planar region of a flow. Current processing techniques offer high precision (1 pct) velocity estimates, but can require several hours of processing time on specialized array processors. Under some circumstances, a simple, fast, less accurate (approx. 5 pct), data reduction technique which also gives unambiguous velocity vector information is acceptable. A direct space domain processing technique was examined. The direct space domain processing technique was found to be far superior to any other techniques known, in achieving the objectives listed above. It employs a new data coding and reduction technique, where the particle time history information is used directly. Further, it has no 180 deg directional ambiguity. A complex convection vortex flow was recorded and completely processed in under 2 minutes on an 80386 based PC, producing a 2-D velocity vector map of the flow field. Hence, using this new space domain vector scanning (VS) technique, pulsed laser velocimetry data can be reduced quickly and reasonably accurately, without specialized array processing hardware.

Wernet, Mark P.↗

Application of high-performance computing to numerical simulation of human movement

We have examined the feasibility of using massively-parallel and vector-processing supercomputers to solve large-scale optimization problems for human movement. Specifically, we compared the computational expense of determining the optimal controls for the single support phase of gait using a conventional serial machine (SGI Iris 4D25), a MIMD parallel machine (Intel iPSC/860), and a parallel-vector-processing machine (Cray Y-MP 8/864). With the human body modeled as a 14 degree-of-freedom linkage actuated by 46 musculotendinous units, computation of the optimal controls for gait could take up to 3 months of CPU time on the Iris. Both the Cray and the Intel are able to reduce this time to practical levels. The optimal solution for gait can be found with about 77 hours of CPU on the Cray and with about 88 hours of CPU on the Intel. Although the overall speeds of the Cray and the Intel were found to be similar, the unique capabilities of each machine are better suited to different portions of the computational algorithm used. The Intel was best suited to computing the derivatives of the performance criterion and the constraints whereas the Cray was best suited to parameter optimization of the controls. These results suggest that the ideal computer architecture for solving very large-scale optimal control problems is a hybrid system in which a vector-processing machine is integrated into the communication network of a MIMD parallel machine.

NASA Discipline Musculoskeletal↗

Large-Scale Numerical Simulations of Human Motion

This paper examines the feasibility of using massively-parallel and vector-processing supercomputers to solve large-scale optimal control problems for human movement. Specifically, we compare the computational expense of determining the optimal controls for the single support phase of walking using a conventional serial machine (a Silicon Graphics Personal Iris 4D25 workstation), a MIMD parallel machine (an Intel iPSC/860 comprising 128 processors), and a parallel-vector-processing machine (a Cray Y-MP 8/864). With the human body modeled as a 14 degree-of-freedom linkage actuated by 46 musculotendinous units, computation of the optimal controls for walking could take up to 3 months of CPU time on the Iris. Both the Cray Y-MP and the Intel iPSC/860 are able to reduce this time to practical levels. The optimal control solution for walking can be found with about 77 hours of CPU time on the Cray, and with about 88 hours of CPU time on the Intel. Although the overall speeds of the Cray and the Intel were found to be similar, the unique capabilities of each machine are best suited to different parts of the optimal control algorithm used. The Intel performed best in the calculation of the derivatives of the performance criterion and the constraints. In contrast, the Cray performed best during parameter optimization of the controls. These results suggest that the ideal computer architecture for solving very large-scale optimal control problems is a hybrid system in which a vector-processing machine is integrated into the communication network of a MIMD parallel machine.

Anderson, Frank C.↗

Numerical analysis of flow through scramjet engine inlets

A set of computer programs was developed to analyze flow through supersonic combustion ramjet (scramjet) inlets. These programs solve either the two or three dimensional Euler/Navier-Stokes equations in full conservation form by MacCormack's explicit or explicit-implicit method. An algebraic two-layer eddy viscosity model is used for turbulent flow calculations. The programs are operational in Control Data CYBER-200 series vector-processing computer system and were optimized to take maximum advantage of the vector processing capability of the system. Since their development, the programs were extensively verified and used to analyze a number of very complex inlet configurations. Results are presented from two-dimensional, quasi-three-dimensional, and three-dimensional analyses of the inlet flow field to illustrate the use of the program.

Kumar, Ajay↗

Numerical analysis of flow through scramjet engine inlets

A set of computer programs has been developed to analyze flow through supersonic combustion ramjet (scramjet) inlets. These programs solve either the two- or three-dimensional Euler/Navier-Stokes equations in full conservation form by MacCormack's explicit or explicit-implicit method. An algebraic two-layer eddy viscosity model is used for turbulent flow calculations. The programs are operational on Control Data CYBER-200 series vector-processing computer system and have been optimized to take maximum advantage of the vector processing capability of the system. Since their development, the programs have been extensively verified and used to analyze a number of very complex inlet configurations. In this paper, results are presented from two-dimensional, quasi-three-dimensional, and three-dimensional analyses of the inlet flow field to illustrate the use of the programs.

Kumar, Ajay↗

Study of the modifications needed for efficient operation of NASTRAN on the Control Data Corporation STAR-100 computer

NASA structural analysis (NASTRAN) computer program is operational on three series of third generation computers. The problem and difficulties involved in adapting NASTRAN to a fourth generation computer, namely, the Control Data STAR-100, are discussed. The salient features which distinguish Control Data STAR-100 from third generation computers are hardware vector processing capability and virtual memory. A feasible method is presented for transferring NASTRAN to Control Data STAR-100 system while retaining much of the machine-independent code. Basic matrix operations are noted for optimization for vector processing.

Source record↗

Modeling material failure with a vectorized routine

The computational aspects of modelling material failure in structural wood members are presented with particular reference to vector processing aspects. Wood members are considered to be highly orthotropic, inhomogeneous, and discontinuous due to the complex microstructure of wood material and the presence of natural growth characteristics such as knots, cracks and cross grain in wood members. The simulation of strength behavior of wood members is accomplished through the use of a special purpose finite element/fracture mechanics routine, program STARW (Strength Analysis Routine for Wood). Program STARW employs quadratic finite elements combined with singular crack tip elements in a finite element mesh. Vector processing techniques are employed in mesh generation, stiffness matrix formation, simultaneous equation solution, and material failure calculations. The paper addresses these techniques along with the time and effort requirements needed to convert existing finite element code to a vectorized version. Comparisons in execution time between vectorized and nonvectorized routines are provided.

Cramer, S. M.↗

Optimizing a Multiple Right-Hand Side Dslash Kernel for Intel Knights Corner

Lattice Quantum Chromo-dynamics (LQCD) stands unique as the only computationally tractable, non-perturbative, and model-independent quantum field theory of the strong nuclear force. LQCD achieves tractability by discretizing our familiar 4-dimensional space-time onto a hypercubic lattice. The core component of the linear solver used in LQCD is the Wilson-Dirac operator whose finite difference component is called the Dslash operator. Dslash is a nearest neighbor stencil operator summing matrix-vector multiplications over lattice points. The arithmetic intensity of a regular Dslash operator acting on a single vector in single precision is between 0.88-2.29 (depending on the tricks employed) making it memory bandwidth bound on most current architectures, including Xeon Phi. Maximizing the performance of Dslash on Xeon Phi requires deliberate exploitation of its architectural features. One key feature of Xeon Phi is its powerful vector processing unit with lengthy vector registers. There is a class of LQCD computations which aims to solve the Dirac equation for the same lattice gauge fields multiple times. In this case one can gain additional reuse of the gauge field, and potentially simplified vectorization, through applying Dslash to several vectors at the same time; often referred to as Multiple Right Hand Side (MRHS) Dslash. Previous work in this area focused on applying this approach to the Highly Improved Staggered operator, and in cases to the multi-Grid algorithm for Wilson clover fermions. Here, we present two approaches to MRHS Dslash: a vector register blocking approach and one using the software package QPhiX with a custom Code Generator for low-level intrinsics. We explore the Knights Corner instruction set, memory hierarchy, and threading capabilities. For standard lattice sizes used in testing, we observed significant speedups, with sustained performance of over 700 GFLOPS (single precision) in one instance. We achieved up to 27% of theoretical peak performance compared to a maximum of 12% obtained by the previous single RHS method in QPhiX.

Joó, Bálint↗

Simplified model of statistically stationary spacecraft rotation and associated induced gravity environments

A stochastic model of spacecraft motion was developed based on the assumption that the net torque vector due to crew activity and rocket thruster firings is a statistically stationary Gaussian vector process. The process had zero ensemble mean value, and the components of the torque vector were mutually stochastically independent. The linearized rigid-body equations of motion were used to derive the autospectral density functions of the components of the spacecraft rotation vector. The cross-spectral density functions of the components of the rotation vector vanish for all frequencies so that the components of rotation were mutually stochastically independent. The autospectral and cross-spectral density functions of the induced gravity environment imparted to scientific apparatus rigidly attached to the spacecraft were calculated from the rotation rate spectral density functions via linearized inertial frame to body-fixed principal axis frame transformation formulae. The induced gravity process was a Gaussian one with zero mean value. Transformation formulae were used to rotate the principal axis body-fixed frame to which the rotation rate and induced gravity vector were referred to a body-fixed frame in which the components of the induced gravity vector were stochastically independent. Rice's theory of exceedances was used to calculate expected exceedance rates of the components of the rotation and induced gravity vector processes.

Fichtl, G. H.↗

Landau Collision Integral Solver with Adaptive Mesh Refinement on Emerging Architectures

The Landau collision integral is an accurate model for the small-angle dominated Coulomb collisions in fusion plasmas. Here, we investigate a high order accurate, fully conservative, finite element discretization of the nonlinear multispecies Landau integral with adaptive mesh refinement using the PETSc library (ulwww.mcs.anl.gov/petsc). We develop algorithms and techniques to efficiently utilize emerging architectures with an approach that minimizes memory usage and movement and is suitable for vector processing. The Landau collision integral is vectorized with Intel AVX-512 intrinsics and the solver sustains as much as 22% of the theoretical peak flop rate of the Second Generation Intel Xeon Phi (``Knights Landing'') processor.

97 MATHEMATICS AND COMPUTING↗

Parallel implementation of an algorithm for Delaunay triangulation

The theory and practice of implementing Tanemura's algorithm for 3D Delaunay triangulation on Intel's Gamma prototype, a 128 processor MIMD computer, is described. Efficient implementation of Tanemura's algorithm on a conventional, vector processing supercomputer is problematic. It does not vectorize to any significant degree and requires indirect addressing. Efficient implementation on a parallel architecture is possible, however. Speeds in excess of 20 times a single processor Cray Y-MP are realized on 128 processors of the Intel Gamma prototype.

Merriam, Marshal L.↗

Vector algorithms for geometrically nonlinear 3D finite element analysis

Algorithms for geometrically nonlinear finite element analysis are presented which exploit the vector processing capability of the VPS-32, which is closely related to the CYBER 205. By manipulating vectors (which are long lists of numbers) rather than individual numbers, very high processing speeds are obtained. Long vector lengths are obtained without extensive replication or reordering by storage of intermediate results in strategic patterns at all stages of the computations. Comparisons of execution times with those from programs using either scalar or other vector programming techniques indicate that the algorithms presented are quite efficient.

Whitcomb, John D.↗

An improved exceedance theory for combined random stresses

An extension is presented of Rice's classic solution for the exceedances of a constant level by a single random process to its counterpart for an n-dimensional vector process. An interaction boundary, analogous to the constant level considered by Rice for the one-dimensional case, is assumed in the form of a hypersurface. The theory for the numbers of boundary exceedances is developed by using a joint statistical approach which fully accounts for all cross-correlation effects. An exact expression is derived for the n-dimensional exceedance density function, which is valid for an arbitrary interaction boundary. For application to biaxial states of combined random stress, the general theory is reduced to the two-dimensional case. An elliptical stress interaction boundary is assumed and the exact expression for the density function is presented. The equations are expressed in a format which facilitates calculating the exceedances by numerically evaluating a line integral. The behavior of the density function for the two-dimensional case is briefly discussed.

Lester, H. C.↗

Feasibility study for a numerical aerodynamic simulation facility. Volume 3: FMP language specification/user manual

The manual is intended to show the revisions and additions to the current STAR FORTRAN. The changes are made to incorporate an FMP (Flow Model Processor) for use in the Numerical Aerodynamic Simulation Facility (NASF) for the purpose of simulating fluid flow over three-dimensional bodies in wind tunnel environments and in free space. The FORTRAN programming language for the STAR-100 computer contains both CDC and unique STAR extensions to the standard FORTRAN. Several of the STAR FORTRAN extensions to standard FOR-TRAN allow the FORTRAN user to exploit the vector processing capabilities of the STAR computer. In STAR FORTRAN, vectors can be expressed with an explicit notation, functions are provided that return vector results, and special call statements enable access to any machine instruction.

Kenner, B. G.↗

Earth Observing System: Present Capabilities and Promises for the Future

The Earth Observing System (EOS) is a space-based observing system comprised of a series of satellite sensors by which scientists can monitor the Earth, a Data and Information System (EOSDIS) enabling researchers worldwide to access the satellite data, and an interdisciplinary science research program to interpret the satellite data. During this year 5 EOS science missions are scheduled for launch, representing observations of (i) total solar irradiance, (ii) Earth radiation budget, (iii) land cover & land use change, (iv) ocean processes (vector wind, sea surface temperature, and ocean color), (v) atmospheric processes (aerosol and cloud properties, water vapor, and temperature and moisture profiles), and (vi) atmospheric chemistry (both tropospheric and stratospheric). In succeeding years many more satellites will be launched that will contribute immeasurably to our understanding of the Earth's environment. In this presentation I will describe how scientists intend to use MODIS, an earth-viewing cross-track scanning spectroradiometer to be launched on the Terra satellite in summer 1999, for the remote sensing of cloud and aerosol properties. MODIS will scan a swath width sufficient to provide nearly complete global coverage every two days from a polar-orbiting, sun-synchronous, platform at an altitude of 705 km. In addition, I will describe key elements of other instruments recently launched or planned for flight in the coming months. Finally, I will lay out a plan for the future space-based observing system being planned by NASA and its partners.

King, Michael D.↗

Clouds and Radiation from the Earth Observing System

The Earth Observing System (EOS) is a space-based observing system comprised of a series of satellite sensors by which scientists can monitor the Earth, a Data and Information System (EOSDIS) enabling researchers worldwide to access the satellite data, and an interdisciplinary science research program to interpret the satellite data. During this year, four EOS science missions are scheduled for launch, representing observations of (i) total solar irradiance, (ii) Earth radiation budget, (iii) land cover & land use change, (iv) ocean processes (vector wind, sea surface temperature, and ocean color), (v) atmospheric processes (aerosol and cloud properties, water vapor, and temperature and moisture profiles), and (vi) tropospheric chemistry. In succeeding years many more satellites will be launched that will contribute immeasurably to our understanding of the Earth's environment. In this presentation we will describe how scientists are using MODIS (Moderate Resolution Imaging Spectroradiometer), an earth-viewing cross-track scanning spectroradiometer on the Terra (AM-1) satellite, for the remote sensing of cloud and aerosol properties. MODIS scans a swath width sufficient to provide nearly complete global coverage every two days from a polar-orbiting, sun-synchronous, platform at an altitude of 705 km, and provides images in 36 spectral bands between 0.415 and 14.235 pm with spatial resolutions of 250 m (2 bands), 500 m (5 bands) and 1000 m (29 bands). These bands have been carefully selected to enable advanced studies of land, ocean, and atmosphere processes.

King, Michael D.↗