SEARCH · Engineering Papers

Results for “compute in memory”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 235 records · Page 13

All sources NASA NTRS DOE OSTI

NASA NTRS1992

Electrostatic Particle-In-Cell Code For Hypercube Computer

Code simulates two-dimensional motions of plasma particles in self-consistent electrostatic field and externally applied magnetic field developed for execution on Mark IIIfp hypercube computer. Based on generalization of one-dimensional particle-in-cell algorithm applicable to many different parallel computing architectures. Intermediate product of continuing effort to speed particle-in-cell computations by taking advantage of distributed-memory parallel computers like those of hypercube class.

Ferraro, Robert D.↗

NASA NTRS1998

First- and Second-Order Sensitivity Analysis of a P-Version Finite Element Equation Via Automatic Differentiation

Sensitivity analysis is a technique for determining derivatives of system responses with respect to design parameters. Among many methods available for sensitivity analysis, automatic differentiation has been proven through many applications in fluid dynamics and structural mechanics to be an accurate and easy method for obtaining derivatives. Nevertheless, the method can be computational expensive and can require a high memory space. This project will apply an automatic differentiation tool, ADIFOR, to a p-version finite element code to obtain first- and second- order then-nal derivatives, respectively. The focus of the study is on the implementation process and the performance of the ADIFOR-enhanced codes for sensitivity analysis in terms of memory requirement, computational efficiency, and accuracy.

Hou, Gene↗

NASA NTRS2005

Parallel Finite Element Domain Decomposition for Structural/Acoustic Analysis

A domain decomposition (DD) formulation for solving sparse linear systems of equations resulting from finite element analysis is presented. The formulation incorporates mixed direct and iterative equation solving strategics and other novel algorithmic ideas that are optimized to take advantage of sparsity and exploit modern computer architecture, such as memory and parallel computing. The most time consuming part of the formulation is identified and the critical roles of direct sparse and iterative solvers within the framework of the formulation are discussed. Experiments on several computer platforms using several complex test matrices are conducted using software based on the formulation. Small-scale structural examples are used to validate thc steps in the formulation and large-scale (l,000,000+ unknowns) duct acoustic examples are used to evaluate the ORIGIN 2000 processors, and a duster of 6 PCs (running under the Windows environment). Statistics show that the formulation is efficient in both sequential and parallel computing environmental and that the formulation is significantly faster and consumes less memory than that based on one of the best available commercialized parallel sparse solvers.

Nguyen, Duc T.↗

NASA NTRS1984

Trends in computational capabilities for fluid dynamics

Milestones in the development of computational aerodynamics are reviewed together with past, present, and future computer performance (speed and memory) trends. Factors influencing computer performance requirements for both steady and unsteady flow simulations are identified. Estimates of computer speed and memory that are required to calculate both inviscid and viscous, steady and unsteady flows about airfoils, wings, and simple wing body configurations are presented and compared to computer performance which is either currently available, or is expected to be available before the end of this decade. Finally, estimates of the amounts of computer time that are required to determine flutter boundaries of airfoils and wings at transonic Mach numbers are presented and discussed.

Peterson, V. L.↗

NASA NTRS1985

Trends in computational capabilities for fluid dynamics

Peterson, V. L.↗

NASA NTRS1993

The alignment-distribution graph

Implementing a data-parallel language such as Fortran 90 on a distributed-memory parallel computer requires distributing aggregate data objects (such as arrays) among the memory modules attached to the processors. The mapping of objects to the machine determines the amount of residual communication needed to bring operands of parallel operations into alignment with each other. We present a program representation called the alignment distribution graph that makes these communication requirements explicit. We describe the details of the representation, show how to model communication cost in this framework, and outline several algorithms for determining object mappings that approximately minimize residual communication.

Chatterjee, Siddhartha↗

NASA NTRS1993

The alignment-distribution graph

Implementing a data-parallel language such as Fortran 90 on a distributed-memory parallel computer requires distributing aggregate data objects (such as arrays) among the memory modules attached to the processors. The mapping of objects to the machine determines the amount of residual communication needed to bring operands of parallel operations into alignment with each other. We present a program representation called the alignment-distribution graph that makes these communication requirements explicit. We describe the details of the representation, show how to model communication cost in this framework, and outline several algorithms for determining object mappings that approximately minimize residual communication.

Chatterjee, Siddhartha↗

NASA NTRS2000

A Two Colorable Fourth Order Compact Difference Scheme and Parallel Iterative Solution of the 3D Convection Diffusion Equation

A new fourth order compact difference scheme for the three dimensional convection diffusion equation with variable coefficients is presented. The novelty of this new difference scheme is that it Only requires 15 grid points and that it can be decoupled with two colors. The entire computational grid can be updated in two parallel subsweeps with the Gauss-Seidel type iterative method. This is compared with the known 19 point fourth order compact differenCe scheme which requires four colors to decouple the computational grid. Numerical results, with multigrid methods implemented on a shared memory parallel computer, are presented to compare the 15 point and the 19 point fourth order compact schemes.

Zhang, Jun↗

NASA NTRS2002

Memory-Intensive Benchmarks: IRAM vs. Cache-Based Machines

The increasing gap between processor and memory performance has lead to new architectural models for memory-intensive applications. In this paper, we explore the performance of a set of memory-intensive benchmarks and use them to compare the performance of conventional cache-based microprocessors to a mixed logic and DRAM processor called VIRAM. The benchmarks are based on problem statements, rather than specific implementations, and in each case we explore the fundamental hardware requirements of the problem, as well as alternative algorithms and data structures that can help expose fine-grained parallelism or simplify memory access patterns. The benchmarks are characterized by their memory access patterns, their basic control structures, and the ratio of computation to memory operation.

Biswas, Rupak↗

NASA NTRS2005

Real-Time Principal-Component Analysis

A recently written computer program implements dominant-element-based gradient descent and dynamic initial learning rate (DOGEDYN), which was described in Method of Real-Time Principal-Component Analysis (NPO-40034) NASA Tech Briefs, Vol. 29, No. 1 (January 2005), page 59. To recapitulate: DOGEDYN is a method of sequential principal-component analysis (PCA) suitable for such applications as data compression and extraction of features from sets of data. In DOGEDYN, input data are represented as a sequence of vectors acquired at sampling times. The learning algorithm in DOGEDYN involves sequential extraction of principal vectors by means of a gradient descent in which only the dominant element is used at each iteration. Each iteration includes updating of elements of a weight matrix by amounts proportional to a dynamic initial learning rate chosen to increase the rate of convergence by compensating for the energy lost through the previous extraction of principal components. In comparison with a prior method of gradient-descent-based sequential PCA, DOGEDYN involves less computation and offers a greater rate of learning convergence. The sequential DOGEDYN computations require less memory than would parallel computations for the same purpose. The DOGEDYN software can be executed on a personal computer.

Duong, Vu↗

NASA NTRS1988

Computer viruses

The worm, Trojan horse, bacterium, and virus are destructive programs that attack information stored in a computer's memory. Virus programs, which propagate by incorporating copies of themselves into other programs, are a growing menace in the late-1980s world of unprotected, networked workstations and personal computers. Limited immunity is offered by memory protection hardware, digitally authenticated object programs,and antibody programs that kill specific viruses. Additional immunity can be gained from the practice of digital hygiene, primarily the refusal to use software from untrusted sources. Full immunity requires attention in a social dimension, the accountability of programmers.

Denning, Peter J.↗

NASA NTRS1993

MEDUSA - An overset grid flow solver for network-based parallel computer systems

Continuing improvement in processing speed has made it feasible to solve the Reynolds-Averaged Navier-Stokes equations for simple three-dimensional flows on advanced workstations. Combining multiple workstations into a network-based heterogeneous parallel computer allows the application of programming principles learned on MIMD (Multiple Instruction Multiple Data) distributed memory parallel computers to the solution of larger problems. An overset-grid flow solution code has been developed which uses a cluster of workstations as a network-based parallel computer. Inter-process communication is provided by the Parallel Virtual Machine (PVM) software. Solution speed equivalent to one-third of a Cray-YMP processor has been achieved from a cluster of nine commonly used engineering workstation processors. Load imbalance and communication overhead are the principal impediments to parallel efficiency in this application.

Smith, Merritt H.↗

NASA NTRS1978

Microprocessor arrays for large scale computation

An important new direction in computer architecture centers around the achievement of very high computational power (capacity, speed and reliability) through the use of tens of thousands of microprocessors, micromemories, and switch modules, all interconnected into a large homogeneous network using one of certain advanced connection schemes. When surrounded and supported by conventional computers and memories, such a machine holds potential for out-performing both conventional and array-based computers of the mid-1980's by one to two orders of magnitude, at least for particular classes of applications amenable to high parallelism, such as aerodynamic simulation. The homogeneous feature of this machine concept also implies size extendability, fault tolerance, and improved flexibility to handle a variety of algorithms of interest. Current work is addressing the design of technologically efficient interconnection configurations and the development of new computation algorithms that are especially efficient for highly parallel computation.

Kautz, W. H.↗

NASA NTRS1997

Computational Issues in Damping Identification for Large Scale Problems

Two damping identification methods are tested for efficiency in large-scale applications. One is an iterative routine, and the other a least squares method. Numerical simulations have been performed on multiple degree-of-freedom models to test the effectiveness of the algorithm and the usefulness of parallel computation for the problems. High Performance Fortran is used to parallelize the algorithm. Tests were performed using the IBM-SP2 at NASA Ames Research Center. The least squares method tested incurs high communication costs, which reduces the benefit of high performance computing. This method's memory requirement grows at a very rapid rate meaning that larger problems can quickly exceed available computer memory. The iterative method's memory requirement grows at a much slower pace and is able to handle problems with 500+ degrees of freedom on a single processor. This method benefits from parallelization, and significant speedup can he seen for problems of 100+ degrees-of-freedom.

Pilkey, Deborah L.↗

NASA NTRS1987

The use of computers for instruction in fluid dynamics

Applications for computers which improve instruction in fluid dynamics are examined. Computers can be used to illustrate three-dimensional flow fields and simple fluid dynamics mechanisms, to solve fluid dynamics problems, and for electronic sketching. The usefulness of computer applications is limited by computer speed, memory, and software and the clarity and field of view of the projected display. Proposed advances in personal computers which will address these limitations are discussed. Long range applications for computers in education are considered.

Watson, Val↗

NASA NTRS2001

Comparison of the NASCAP/GEO, POLAR, SEE Charging Handbook, and NASCAP-2K.1 Spacecraft Charging Codes

The NASA Charging Analyzer Program (NASCAP) spacecraft charging software developed by Maxwell Technologies has been widely used for the past fifteen to twenty years in satellite design and investigation of spacecraft charging related anomalies. Individual versions of the NASCAP software are available for use in low inclination, low Earth orbit environments (NASCAP[LEO) and geostationary orbit environments (NASCAP/GEO). In addition, the Potentials of Large objects in the Auroral Region (POLAR) code is available for use in LEO polar orbit environments. NASCAP/GEO and POLAR were both written in the 1980's using algorithms appropriate for the computers of the time. They solve the Poisson-Vlasov system for currents and densities assuming limited speed and memory of computer systems standard for the day. In addition, use of the charging models requires individual input files that are not readily transported into the various codes to facilitate comparison of results by the user.

Neergaard, Linda E.↗

NASA NTRS1972

Coordination and establishment of centralized facilities and services for the University of Alaska ERTS survey of the Alaskan environment

The author has identified the following significant results. Specifications have been prepared for the engineering design and construction of a digital color display unit which will be used for automatic processing of ERTS data. The color display unit is a disk refresh memory with computer interfaced input and a color cathode ray tube output display. The system features both analog and digital post disk data manipulation and a versatile color coding device suitable for displaying not only images, but also computer generated graphics such as diagrams, maps, and overlays. Input is from IBM compatible 9 track, 800 BPI tapes, as generated by an IBM 360 computer. ERTS digital tapes are read into the 360, where various analyses such as maximum likelihood classification are performed and the results are written on a magnetic tape which is the input to the color display unit. The greatest versatility in the data manipulation area is provided by the minicomputer built into the color display unit, which is off-line from the main 360 computer. The minicomputer is able to read any line from the refresh disk and place it in its 4K-16 bit memory. Considerable flexibility is available for post-processing enhancement of images by the investigator.

Belon, A. E.↗

NASA NTRS1990

Calibration For On-Machine Inspections

Permanent calibration setup on numerically-controlled machine tool enables fast and reliable calibration of automated probes for inspection of workpieces while still mounted on machine. Setup includes one or more artifacts, dimensions and locations known precisely and stored in memory of computer controlling machine. Before probes used to inspect newly machined part, used to check artifacts. Probe measurements compared with values in memory to determine whether probe readings need adjustment.

Haymaker, Carlton L., Jr.↗