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 181 records · Page 10

All sources NASA NTRS DOE OSTI

NASA NTRS2011

A Multiple Sphere T-Matrix Fortran Code for Use on Parallel Computer Clusters

A general-purpose Fortran-90 code for calculation of the electromagnetic scattering and absorption properties of multiple sphere clusters is described. The code can calculate the efficiency factors and scattering matrix elements of the cluster for either fixed or random orientation with respect to the incident beam and for plane wave or localized- approximation Gaussian incident fields. In addition, the code can calculate maps of the electric field both interior and exterior to the spheres.The code is written with message passing interface instructions to enable the use on distributed memory compute clusters, and for such platforms the code can make feasible the calculation of absorption, scattering, and general EM characteristics of systems containing several thousand spheres.

Mackowski, D. W.↗

DOE OSTI2025

Angular-spatial hp -adaptivity for radiative transfer with discontinuous Galerkin spectral element methods

Radiative transfer is important for many science and engineering applications, and numerical simulations of radiative transfer can be challenging. For instance, the radiation field is seven-dimensional – three spatial, two angular, one wavelength, and one temporal – and often features steep gradients. Therefore, memory usage is a key issue. To reduce memory, some past work has investigated the use of adaptive mesh refinement (AMR), typically for either the spatial or angular coordinate, and typically for only h -adaptivity. Here, we propose the use of AMR for the spatial and angular coordinates together, and the use of h - and p -adaptivity together as hp -AMR for the potential for further memory savings. We implemented the proposed method for several test cases in two spatial and one angular dimension, with the discontinuous Galerkin spectral element method. These test cases featured highly anisotropic angular radiation, with or without steep spatial gradients. Our primary findings from these test cases were: (1) Angular hp -adaptivity can deliver the radiation solution with the same accuracy as, and with much less computational memory than, uniform angular h - or p -refinements, or angular h -adaptivity alone. This is most obvious when the incoming radiation is highly anisotropic, in which case the savings can be orders of magnitude. (2) Full spatial-angular hp -adaptivity is more efficient in solution representation, compared to solely spatial or solely angular -adaptivity. This is most evident when steep gradients are present in both the spatial and angular distribution. These results suggest that adaptive spatial- hp angular-refinement may perform well in large-scale seven-dimensional applications.

Adaptive refinement↗

NASA NTRS1992

Using adaptive grid in modeling rocket nozzle flow

The mechanical behavior of a rocket motor internal flow field results in a system of nonlinear partial differential equations which cannot be solved analytically. However, this system of equations called the Navier-Stokes equations can be solved numerically. The accuracy and the convergence of the solution of the system of equations will depend largely on how precisely the sharp gradients in the domain of interest can be resolved. With the advances in computer technology, more sophisticated algorithms are available to improve the accuracy and convergence of the solutions. An adaptive grid generation is one of the schemes which can be incorporated into the algorithm to enhance the capability of numerical modeling. It is equivalent to putting intelligence into the algorithm to optimize the use of computer memory. With this scheme, the finite difference domain of the flow field called the grid does neither have to be very fine nor strategically placed at the location of sharp gradients. The grid is self adapting as the solution evolves. This scheme significantly improves the methodology of solving flow problems in rocket nozzles by taking the refinement part of grid generation out of the hands of computational fluid dynamics (CFD) specialists and place it into the computer algorithm itself.

Chow, Alan S.↗

NASA NTRS2006

Applying Parallel Adaptive Methods with GeoFEST/PYRAMID to Simulate Earth Surface Crustal Dynamics

This viewgraph presentation reviews the use Adaptive Mesh Refinement (AMR) in simulating the Crustal Dynamics of Earth's Surface. AMR simultaneously improves solution quality, time to solution, and computer memory requirements when compared to generating/running on a globally fine mesh. The use of AMR in simulating the dynamics of the Earth's Surface is spurred by future proposed NASA missions, such as InSAR for Earth surface deformation and other measurements. These missions will require support for large-scale adaptive numerical methods using AMR to model observations. AMR was chosen because it has been successful in computation fluid dynamics for predictive simulation of complex flows around complex structures.

GeoFEST↗

DOE OSTI2022

Understanding the Impact of Data Staging for Coupled Scientific Workflows

We report the rate of data generated by cutting-edge experimental science facilities and large-scale simulations enabled by current high-performance computing (HPC) systems has continued to grow at a far greater pace than the development of the network and storage capabilities on which these systems rely. To cope with this challenge, scientist are moving toward the creation of autonomous experiments and HPC simulations using machine learning. However, efficiently moving, storing, and processing large amounts of data away from the point of origin presents an incredible challenge. In-memory computing, in situ analysis, data staging, and data streaming are recognized viable alternatives to traditional file-based methods for transferring data between coupled workflows. However, the performance trade-offs and limitations for these methods are not fully understood when used in HPC applications. This article presents a comprehensive performance assessment of the current solutions for data staging when applied to applications that are not necessary I/O intensive which makes them not ideal candidates for these methods. Our study is based on experiments running at scale on Oak Ridge National Laboratory's Summit supercomputer using applications and simulations that cover typical computational motifs and patterns. We investigated the usability and cost/benefit trade-offs of staging algorithms for HPC applications under different scenarios and highlight opportunities for optimizing the dataflow between coupled simulation workflows.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS2003

Software for Acquiring Image Data for PIV

PIV Acquisition (PIVACQ) is a computer program for acquisition of data for particle-image velocimetry (PIV). In the PIV system for which PIVACQ was developed, small particles entrained in a flow are illuminated with a sheet of light from a pulsed laser. The illuminated region is monitored by a charge-coupled-device camera that operates in conjunction with a data-acquisition system that includes a frame grabber and a counter-timer board, both installed in a single computer. The camera operates in "frame-straddle" mode where a pair of images can be obtained closely spaced in time (on the order of microseconds). The frame grabber acquires image data from the camera and stores the data in the computer memory. The counter/timer board triggers the camera and synchronizes the pulsing of the laser with acquisition of data from the camera. PIVPROC coordinates all of these functions and provides a graphical user interface, through which the user can control the PIV data-acquisition system. PIVACQ enables the user to acquire a sequence of single-exposure images, display the images, process the images, and then save the images to the computer hard drive. PIVACQ works in conjunction with the PIVPROC program which processes the images of particles into the velocity field in the illuminated plane.

Wernet, Mark P.↗

NASA NTRS2003

Improved Collision-Detection Method for Robotic Manipulator

An improved method has been devised for the computational prediction of a collision between (1) a robotic manipulator and (2) another part of the robot or an external object in the vicinity of the robot. The method is intended to be used to test commanded manipulator trajectories in advance so that execution of the commands can be stopped before damage is done. The method involves utilization of both (1) mathematical models of the robot and its environment constructed manually prior to operation and (2) similar models constructed automatically from sensory data acquired during operation. The representation of objects in this method is simpler and more efficient (with respect to both computation time and computer memory), relative to the representations used in most prior methods. The present method was developed especially for use on a robotic land vehicle (rover) equipped with a manipulator arm and a vision system that includes stereoscopic electronic cameras. In this method, objects are represented and collisions detected by use of a previously developed technique known in the art as the method of oriented bounding boxes (OBBs). As the name of this technique indicates, an object is represented approximately, for computational purposes, by a box that encloses its outer boundary. Because many parts of a robotic manipulator are cylindrical, the OBB method has been extended in this method to enable the approximate representation of cylindrical parts by use of octagonal or other multiple-OBB assemblies denoted oriented bounding prisms (OBPs), as in the example of Figure 1. Unlike prior methods, the OBB/OBP method does not require any divisions or transcendental functions; this feature leads to greater robustness and numerical accuracy. The OBB/OBP method was selected for incorporation into the present method because it offers the best compromise between accuracy on the one hand and computational efficiency (and thus computational speed) on the other hand.

Leger, Chris↗

DOE OSTI2020

An Analysis of System Balance and Architectural Trends Based on Top500 Supercomputers

Supercomputer design is a complex, multi-dimensional optimization process, wherein several subsystems need to be reconciled to meet a desired figure of merit performance for a portfolio of applications and a budget constraint. However, overall, the HPC community has been gravitating towards ever more FLOPS, at the expense of many other subsystems. To draw attention to overall system balance, in this paper, we analyze balance ratios and architectural trends in the world's most powerful supercomputers. Specifically, we have collected performance characteristics of systems between1993 and 2018 based on the Top500 lists, and then analyzed their architectures from diverse system design perspectives. Notably, our analysis studies the performance balance of the machines, across a variety of subsystems such as compute, memory, I/O, interconnect, intra-node connectivity and power. Our analysis reveals that balance ratios of the various subsystems need to be considered carefully alongside the application workload portfolio to provision the subsystem capacity and bandwidth specifications, which can help achieve optimal performance.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS1998

Parallel Implicit Algorithms for CFD

The main goal of this project was efficient distributed parallel and workstation cluster implementations of Newton-Krylov-Schwarz (NKS) solvers for implicit Computational Fluid Dynamics (CFD.) "Newton" refers to a quadratically convergent nonlinear iteration using gradient information based on the true residual, "Krylov" to an inner linear iteration that accesses the Jacobian matrix only through highly parallelizable sparse matrix-vector products, and "Schwarz" to a domain decomposition form of preconditioning the inner Krylov iterations with primarily neighbor-only exchange of data between the processors. Prior experience has established that Newton-Krylov methods are competitive solvers in the CFD context and that Krylov-Schwarz methods port well to distributed memory computers. The combination of the techniques into Newton-Krylov-Schwarz was implemented on 2D and 3D unstructured Euler codes on the parallel testbeds that used to be at LaRC and on several other parallel computers operated by other agencies or made available by the vendors. Early implementations were made directly in Massively Parallel Integration (MPI) with parallel solvers we adapted from legacy NASA codes and enhanced for full NKS functionality. Later implementations were made in the framework of the PETSC library from Argonne National Laboratory, which now includes pseudo-transient continuation Newton-Krylov-Schwarz solver capability (as a result of demands we made upon PETSC during our early porting experiences). A secondary project pursued with funding from this contract was parallel implicit solvers in acoustics, specifically in the Helmholtz formulation. A 2D acoustic inverse problem has been solved in parallel within the PETSC framework.

Keyes, David E.↗

NASA NTRS1986

Transonic Navier-Stokes wing solutions using a zonal approach. Part 2: High angle-of-attack simulation

A computer code is under development whereby the thin-layer Reynolds-averaged Navier-Stokes equations are to be applied to realistic fighter aircraft configurations. This transonic Navier-Stokes code (TNS) utilizes a zonal approach in order to treat complex geometries and satisfy in-core computer memory constraints. The zonal approach was applied to isolated wing geometries in order to facilitate code development. The TNS finite difference algorithm, zonal methodology, and code validation with experimental data is addressed. Also addressed are some numerical issues such as code robustness, efficiency, and accuracy at high angles of attack. Special free-stream-preserving metrics proved an effective way to treat H-mesh singularities over a large range of severe flow conditions, including strong leading edge flow gradients, massive shock induced separation, and stall. Furthermore, lift and drag coefficients were computed for a wing up through CLmax. Numerical oil flow patterns and particle trajectories are presented both for subcritical and transonic flow. These flow simulations are rich with complex separated flow physics and demonstrate the efficiency and robustness of the zonal approach.

Chaderjian, Neal M.↗

NASA NTRS1988

Numerical solution methods for viscoelastic orthotropic materials

Numerical solution methods for viscoelastic orthotropic materials, specifically fiber reinforced composite materials, are examined. The methods include classical lamination theory using time increments, direction solution of the Volterra Integral, Zienkiewicz's linear Prony series method, and a new method called Nonlinear Differential Equation Method (NDEM) which uses a nonlinear Prony series. The criteria used for comparison of the various methods include the stability of the solution technique, time step size stability, computer solution time length, and computer memory storage. The Volterra Integral allowed the implementation of higher order solution techniques but had difficulties solving singular and weakly singular compliance function. The Zienkiewicz solution technique, which requires the viscoelastic response to be modeled by a Prony series, works well for linear viscoelastic isotropic materials and small time steps. The new method, NDEM, uses a modified Prony series which allows nonlinear stress effects to be included and can be used with orthotropic nonlinear viscoelastic materials. The NDEM technique is shown to be accurate and stable for both linear and nonlinear conditions with minimal computer time.

Gramoll, K. C.↗

NASA NTRS1986

Transonic Navier-Stokes wing solutions using a zonal approach. Part 2: High angle-of-attack simulation

A computer code is under development whereby the thin-layer Reynolds-averaged Navier-Stokes equations are to be applied to realistic fighter-aircraft configurations. This transonic Navier-Stokes code (TNS) utilizes a zonal approach in order to treat complex geometries and satisfy in-core computer memory constraints. The zonal approach has been applied to isolated wing geometries in order to facilitate code development. Part 1 of this paper addresses the TNS finite-difference algorithm, zonal methodology, and code validation with experimental data. Part 2 of this paper addresses some numerical issues such as code robustness, efficiency, and accuracy at high angles of attack. Special free-stream-preserving metrics proved an effective way to treat H-mesh singularities over a large range of severe flow conditions, including strong leading-edge flow gradients, massive shock-induced separation, and stall. Furthermore, lift and drag coefficients have been computed for a wing up through CLmax. Numerical oil flow patterns and particle trajectories are presented both for subcritical and transonic flow. These flow simulations are rich with complex separated flow physics and demonstrate the efficiency and robustness of the zonal approach.

Chaderjian, N. M.↗

NASA NTRS1995

Weaves as an Interconnection Fabric for ASIM's and Nanosatellites

Many of the micromachines under consideration require computer support, indeed, one of the appeals of this technology is the ability to intermix mechanical, optical, analog, and digital devices on the same substrate. The amount of computer power is rarely an issue, the sticking point is the complexity of the software required to make effective use of these devices. Micromachines are the nano-technologist's equivalent of 'golden screws'. In other words, they will be piece parts in larger assemblages. For example, a nano-satellite may be composed of stacked silicon wafers where each wafer contains hundreds to thousands of micromachines, digital controllers, general purpose computers, memories, and high-speed bus interconnects. Comparatively few of these devices will be custom designed, most will be stock parts selected from libraries and catalogs. The novelty will lie in the interconnections. For example, a digital accelerometer may be a component part in an adaptive suspension, a monitoring element embedded in the wrapper of a package, or a portion of the smart skin of a launch vehicle. In each case, this device must inter-operate with other devices and probes for the purposes of command, control, and communication. We propose a software technology called 'weaves' that will permit large collections of micromachines and their attendant computers to freely intercommunicate while preserving modularity, transparency, and flexibility. Weaves are composed of networks of communicating software components. The network, and the components comprising it, may be changed even while the software, and the devices it controls, are executing. This unusual degree of software plasticity permits micromachines to dynamically adapt the software to changing conditions and allows system engineers to rapidly and inexpensively develop special purpose software by assembling stock software components in custom configurations.

Gorlick, Michael M.↗

NASA NTRS1987

Tracking 3-D body motion for docking and robot control

An advanced method of tracking three-dimensional motion of bodies has been developed. This system has the potential to dynamically characterize machine and other structural motion, even in the presence of structural flexibility, thus facilitating closed loop structural motion control. The system's operation is based on the concept that the intersection of three planes defines a point. Three rotating planes of laser light, fixed and moving photovoltaic diode targets, and a pipe-lined architecture of analog and digital electronics are used to locate multiple targets whose number is only limited by available computer memory. Data collection rates are a function of the laser scan rotation speed and are currently selectable up to 480 Hz. The tested performance on a preliminary prototype designed for 0.1 in accuracy (for tracking human motion) at a 480 Hz data rate includes a worst case resolution of 0.8 mm (0.03 inches), a repeatability of plus or minus 0.635 mm (plus or minus 0.025 inches), and an absolute accuracy of plus or minus 2.0 mm (plus or minus 0.08 inches) within an eight cubic meter volume with all results applicable at the 95 percent level of confidence along each coordinate region. The full six degrees of freedom of a body can be computed by attaching three or more target detectors to the body of interest.

Donath, M.↗

DOE OSTI2022

Design and Challenges of Edge Computing ASICs on Front-End Electronics

In situ or hardware-embedded data processing of raw signals, close to their source, in radiation detectors is expected to provide dramatic improvements in data quality and volumes. However, the implementation of artificial neural networks (ANNs) in the front-end electronics, and the design of custom integrated circuits (ASICs), comes with challenges. In addition, detectors have to operate with limited power budget and implement complex functionalities in a very dense space. They often are exposed to extreme conditions as they work in high-radiation environments and/or cryogenic temperatures. This paper presents examples of applications and design methodologies for in-situ ANNs, along with the challenges of retaining the fidelity of the trained networks. For illustration, we use the problem of estimating the energy deposited by the radiation from digitized waveforms. The proposed implementation starts with an ML algorithm trained in Qkeras and eventually leads to an equivalent ASIC implementation. The associated design challenges in realizing energy and area efficient implementations in CMOS processes are reviewed. Novel approaches that employ hybrid technologies (combination of CMOS with memristors), in-memory computing models, and bio-inspired spiking neural networks are also highlighted

47 OTHER INSTRUMENTATION↗

NASA NTRS2002

Gilgamesh: A Multithreaded Processor-In-Memory Architecture for Petaflops Computing

Processor-in-Memory (PIM) architectures avoid the von Neumann bottleneck in conventional machines by integrating high-density DRAM and CMOS logic on the same chip. Parallel systems based on this new technology are expected to provide higher scalability, adaptability, robustness, fault tolerance and lower power consumption than current MPPs or commodity clusters. In this paper we describe the design of Gilgamesh, a PIM-based massively parallel architecture, and elements of its execution model. Gilgamesh extends existing PIM capabilities by incorporating advanced mechanisms for virtualizing tasks and data and providing adaptive resource management for load balancing and latency tolerance. The Gilgamesh execution model is based on macroservers, a middleware layer which supports object-based runtime management of data and threads allowing explicit and dynamic control of locality and load balancing. The paper concludes with a discussion of related research activities and an outlook to future work.

management locality load balance↗

DOE OSTI2022

EvaluateFlux, VARPOW, VARPEAK, and DIF3DtoVTK Utility Programs for DIF3D

The DIF3D-VARIANT solver is part of the DIF3D code which was developed at ANL for engineering level predictive analysis of nuclear reactor cores. Over the course of the last 20 years, the DIF3D-VARIANT option of DIF3D has become increasingly more used because of improvements in computer memory capabilities and performance. This is beneficial to the users of DIF3D as DIF3D-VARIANT is able to apply both p- and h- spatial mesh refinement strategies in diffusion and transport theory. Given its greater use, utility programs were created to assist in post-processing the DIF3D output. This document discusses the basic execution process of DIF3D and the output files it creates but is primarily focused on being the manual for the utility programs EvaluateFlux, VARPOW, VARPEAK, and DIF3DtoVTK.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

DOE OSTI2022

Proton Tunable Analog Transistor for Low Power Computing

This project was broadly motivated by the need for new hardware that can process information such as images and sounds right at the point of where the information is sensed (e.g. edge computing). The project was further motivated by recent discoveries by group demonstrating that while certain organic polymer blends can be used to fabricate elements of such hardware, the need to mix ionic and electronic conducting phases imposed limits on performance, dimensional scalability and the degree of fundamental understanding of how such devices operated. As an alternative to blended polymers containing distinct ionic and electronic conducting phases, in this LDRD project we have discovered that a family of mixed valence coordination compounds called Prussian blue analogue (PBAs), with an open framework structure and ability to conduct both ionic and electronic charge, can be used for inkjet-printed flexible artificial synapses that reversibly switch conductance by more than four orders of magnitude based on electrochemically tunable oxidation state. Retention of programmed states is improved by nearly two orders of magnitude compared to the extensively studied organic polymers, thus enabling in-memory compute and avoiding energy costly off-chip access during training. We demonstrate dopamine detection using PBA synapses and biocompatibility with living neurons, evoking prospective application for brain - computer interfacing. By application of electron transfer theory to in-situ spectroscopic probing of intervalence charge transfer, we elucidate a switching mechanism whereby the degree of mixed valency between N-coordinated Ru sites controls the carrier concentration and mobility, as supported by density functional theory (DFT) .

97 MATHEMATICS AND COMPUTING↗