Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “computer hardware”

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 1,639 records · Page 91

Fault Diagnosis of Hybrid Computing Systems Using Chaotic-Map Method

Computing systems are becoming increasingly complex with nodes consisting of a combination of multi-core central processing units (CPUs), many integrated core (MIC) and graphics processing unit (GPU) accelerators. These computing units and their interconnections are subject to different classes of hardware and software faults, which should be detected to support mitigation measures. We present the chaotic-map method that uses the exponential divergence and wide Fourier properties of the trajectories, combined with memory allocations and assignments to diagnose component-level faults in these hybrid computing systems. We propose lightweight codes that utilize highly parallel chaotic-map computations tailored to isolate faults in arithmetic units, memory elements and interconnects. The diagnosis module on a node utilizes pthreads to place chaotic-map threads on CPU and MIC cores, and CUDA C and OpenCL kernels on GPU blocks. We present experimental diagnosis results on five multi-core CPUs; one MIC; and, seven GPUs with typical diagnosis run-times under a minute.

Rao, Nageswara↗

Proxy Applications for Converged Workloads: DMC LDRD Initiative

Modern scientific applications are complicated and require coordination of several components. Proxy application driven software-hardware co-design plays a vital role in driving innovation among the developments of applications, software infrastructure and hardware architecture. Proxy applications are self-contained and simplified codes that are intended to model the performance-critical computations within applications. Applications executing on modern High Performance Computing (HPC) systems are susceptible to network congestion, insufficient memory bandwidth within and across compute nodes, and inadvertent loss of performance due to bugs and unoptimized programming models. Modern numerical simulations and machine learning models play a critical role in studying physical phenomenon under myriad uncertainties. Such applications often exhibit irregular computation and memory accesses at specific regions of the application code, which can contribute to various performance bottlenecks at scale. To mitigate such issues and prepare the next generation hardware for a variety of computation and data movement contingencies, a well-known practice is to consider "proxy" applications as representative motifs for various classes of scientific applications. While there is disagreement in the HPC community on the mechanisms of construction of the proxy applications, there is a strong consensus on their positive impact in co-design. Proxy Applications for Converged Workloads (PACER) is about facilitating software-hardware co-design through proxy applications with the goal of improving the performance of converged science workflows on heterogeneous systems.

97 MATHEMATICS AND COMPUTING↗

Optimal size of the block in block GMRES on GPUs: computational model and experiments

The block version of GMRES (BGMRES) is most advantageous over the single right hand side (RHS) counterpart when the cost of communication is high while the cost of floating point operations is not. This is the particular case on modern graphics processing units (GPUs), while it is generally not the case on traditional central processing units (CPUs). Here, in this paper, experiments on both GPUs and CPUs are shown that compare the performance of BGMRES against GMRES as the number of RHS increases, with a particular focus on GPU performance. The experiments indicate that there are many cases in which BGMRES is slower than GMRES on CPUs, but faster on GPUs. Furthermore, when varying the number of RHS on the GPU, there is an optimal number of RHS where BGMRES is clearly most advantageous over GMRES. A computational model for the GPU is developed using hardware specific parameters, providing insight towards how the qualitative behavior of BGMRES changes as the number of RHS increase, and this model also helps explain the phenomena observed in the experiments.

97 MATHEMATICS AND COMPUTING↗

Stabilizing Inverter-Based Transmission Systems: Power Hardware-in-the-Loop Experiments with a Megawatt-Scale Grid-Forming Inverter

This article presents what the authors believe to be the first experimental verification of the ability of grid-forming (GFM) inverters to stabilize a transmission electric power system that is otherwise unstable. The experiments described here were performed using power hardware-in-the-loop (PHIL) simulation to connect a megawatt-scale battery inverter to a real-time electromagnetic transient (EMT) simulation of the near-future Maui power system. This allows the dynamic interactions between the inverter and the power system to be observed without putting the real power system at risk. The ability to use the actual inverter hardware removes the need to rely on a computer model approximation of the inverter's behavior.

electromagnetic transient simulations↗

Fast b -tagging at the high-level trigger of the ATLAS experiment in LHC Run 3

The ATLAS experiment relies on real-time hadronic jet reconstruction and b-tagging to record fully hadronic events containing b-jets. These algorithms require track reconstruction, which is computationally expensive and could overwhelm the high-level-trigger farm, even at the reduced event rate that passes the ATLAS first stage hardware-based trigger. In LHC Run 3, ATLAS has mitigated these computational demands by introducing a fast neural-network-based b-tagger, which acts as a low-precision filter using input from hadronic jets and tracks. It runs after a hardware trigger and before the remaining high-level-trigger reconstruction. This design relies on the negligible cost of neural-network inference as compared to track reconstruction, and the cost reduction from limiting tracking to specific regions of the detector. In the case of Standard Model HH → bb̅bb̅, a key signature relying on b-jet triggers, the filter lowers the input rate to the remaining high-level trigger by a factor of five at the small cost of reducing the overall signal efficiency by roughly 2%.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Co-design for Particle Applications at Exascale

Co-design across the Exascale Computing Project (ECP) has been critical for both enabling science applications and bringing disparate communities together. Developing and porting applications to the various high-performance computing (HPC) architectures on pre-exascale and exascale computers has been quite challenging due to the diversity of hardware features and software stacks. The Co-design Center for Particle Applications (CoPA) has developed and enhanced the Cabana and PROGRESS/BML libraries to facilitate the creation of new particle applications, make existing particle applications exascale capable, and allow teams to explore new capabilities. Particle methods from atomistic, mesoscale, continuum, through cosmological scales have been built with Cabana, along with new possibilities for application coupling. Similarly, the PROGRESS/BML library has enabled quantum particle applications with linear algebra solvers to use advanced hardware. Across these CoPA-developed libraries, the co-design abstraction layer combines performance portability with math library support to facilitate separation of concerns and directly support science runs.

97 MATHEMATICS AND COMPUTING↗

Triaxial Probe Magnetic Data Analysis

The Triaxial Magnetic Moment Analysis software uses measured magnetic field test data to compute dipole and quadrupole moment information from a hardware element. It is used to support JPL projects needing magnetic control and an understanding of the spacecraft-generated magnetic fields. Evaluation of the magnetic moment of an object consists of three steps: acquisition, conditioning, and analysis. This version of existing software was extensively rewritten for easier data acquisition, data analysis, and report presentation, including immediate feedback to the test operator during data acquisition. While prior JPL computer codes provided the same data content, this program has a better graphic display including original data overlaid with reconstructed results to show goodness of fit accuracy and better appearance of the report graphic page. Data are acquired using three magnetometers and two rotations of the device under test. A clean acquisition user interface presents required numeric data and graphic summaries, and the analysis module yields the best fit (least squares) for the magnetic dipole and/or quadrupole moment of a device. The acquisition module allows the user to record multiple data sets, selecting the best data to analyze, and is repeated three times for each of the z-axial and y-axial rotations. In this update, the y-axial rotation starting position has been changed to an option, allowing either the x- or z-axis to point towards the magnetometer. The code has been rewritten to use three simultaneous axes of magnetic data (three probes), now using two "rotations" of the device under test rather than the previous three rotations, thus reducing handling activities on the device under test. The present version of the software gathers data in one-degree increments, which permits much better accuracy of the fit ted data than the coarser data acquisition of the prior software. The data-conditioning module provides a clean data set for the analysis module. For multiple measurements at a given degree, the first measurement is used. For omitted measurements, the missing field is estimated by linear interpolation between the two nearest measurements. The analysis module was rewritten for the dual rotation, triaxial probe measurement process and now has better moment estimation accuracy, based on the finer one degree of data acquisition resolution. The magnetic moments thus computed are used as an input to summarize the total spacecraft field.

Shultz, Kimberly↗

Error Mitigation of Point-to-Point Communication for Fault-Tolerant Computing

Fault tolerant systems require the ability to detect and recover from physical damage caused by the hardware s environment, faulty connectors, and system degradation over time. This ability applies to military, space, and industrial computing applications. The integrity of Point-to-Point (P2P) communication, between two microcontrollers for example, is an essential part of fault tolerant computing systems. In this paper, different methods of fault detection and recovery are presented and analyzed.

Akamine, Robert L.↗

Time-Accurate Unsteady Pressure Loads Simulated for the Space Launch System at Wind Tunnel Conditions

A transonic flow field about a Space Launch System (SLS) configuration was simulated with the Fully Unstructured Three-Dimensional (FUN3D) computational fluid dynamics (CFD) code at wind tunnel conditions. Unsteady, time-accurate computations were performed using second-order Delayed Detached Eddy Simulation (DDES) for up to 1.5 physical seconds. The surface pressure time history was collected at 619 locations, 169 of which matched locations on a 2.5 percent wind tunnel model that was tested in the 11 ft. x 11 ft. test section of the NASA Ames Research Center's Unitary Plan Wind Tunnel. Comparisons between computation and experiment showed that the peak surface pressure RMS level occurs behind the forward attach hardware, and good agreement for frequency and power was obtained in this region. Computational domain, grid resolution, and time step sensitivity studies were performed. These included an investigation of pseudo-time sub-iteration convergence. Using these sensitivity studies and experimental data comparisons, a set of best practices to date have been established for FUN3D simulations for SLS launch vehicle analysis. To the author's knowledge, this is the first time DDES has been used in a systematic approach and establish simulation time needed, to analyze unsteady pressure loads on a space launch vehicle such as the NASA SLS.

Alter, Stephen J.↗

Enabling NVM for Data-Intensive Scientific Services

Specialized, transient data services are playing an increasingly prominent role in data-intensive scientific computing. These services offer flexible, on-demand pairing of applications with storage hardware using semantics that are optimized for the problem domain. Concurrent with this trend, upcoming scientific computing and big data systems will be deployed with emerging NVM technology to achieve the highest possible price/productivity ratio. Clearly, therefore, we must develop techniques to facilitate the confluence of specialized data services and NVM technology. In this work we explore how to enable the composition of NVM resources within transient distributed services while still retaining their essential performance characteristics. Our approach involves eschewing the conventional distributed file system model and instead projecting NVM devices as remote microservices that leverage user-level threads, RPC services, RMA-enabled network transports, and persistent memory libraries in order to maximize performance. We describe a prototype system that incorporates these concepts, evaluate its performance for key workloads on an exemplar system, and discuss how the system can be leveraged as a component of future data-intensive architectures.

Carns, Philip↗

Cloud computing security.

Cloud computing is a paradigm rapidly being embraced by government and industry as a solution for cost-savings, scalability, and collaboration. While a multitude of applications and services are available commercially for cloud-based solutions, research in this area has yet to fully embrace the full spectrum of potential challenges facing cloud computing. This tutorial aims to provide researchers with a fundamental understanding of cloud computing, with the goals of identifying a broad range of potential research topics, and inspiring a new surge in research to address current issues. We will also discuss real implementations of research-oriented cloud computing systems for both academia and government, including configuration options, hardware issues, challenges, and solutions.

99 GENERAL AND MISCELLANEOUS↗

Control of an artificial infrared environment to simulate complex, time varying orbital conditions

A computer-controlled infrared simulation system has recently been developed for thermal vacuum ground testing of spacecraft in the Space Environment Simulation Laboratory of the NASA Manned Spacecraft Center. The system can simultaneously follow as many as 30 separate time-varying intensity profiles for irradiation of different areas of a test spacecraft, producing an environment that more closely resembles the changing conditions experienced by an orbiting spacecraft. A description of the hardware is included, but emphasis is placed upon computer techniques that have been developed to maximize the capabilities of the control system within the constraints imposed by a limited computer capability. Among the topics discussed are open-loop control, closed-loop control, manual alteration of profiles, compensation for slow response of the radiation source, and compensation for interacting control zones. Examples of typical results obtained using the control system are also presented.

Dewey, R. L.↗

Application of software technology to a future spacecraft computer design

A study was conducted to determine how major improvements in spacecraft computer systems can be obtained from recent advances in hardware and software technology. Investigations into integrated circuit technology indicated that the CMOS/SOS chip set being developed for the Air Force Avionics Laboratory at Wright Patterson had the best potential for improving the performance of spaceborne computer systems. An integral part of the chip set is the bit slice arithmetic and logic unit. The flexibility allowed by microprogramming, combined with the software investigations, led to the specification of a baseline architecture and instruction set.

Labaugh, R. J.↗

Umpire: Application-Focused Management and Coordination of Complex Hierarchical Memory

Advanced architectures like Sierra provide a wide range of memory resources that must often be carefully controlled by the user. These resources have varying capacities, access timing rules, and visibility to different compute resources. Applications must intelligently allocate data in these spaces, and depending on the total amount of memory required, applications may also be forced to move data between different parts of the memory hierarchy. Finally, applications using multiple packages must coordinate effectively to ensure that each package can use the memory resources it needs. To address these challenges, we present Umpire, an application-oriented library for managing memory resources. Specifically, Umpire provides support for querying memory resources, provisioning and allocating memory, and memory introspection. It allows computer scientists and computational physicists to efficiently program the memory hierarchies of current and future high-performance computing architectures, without tying their application to specific hardware or software. In this paper, we describe the design and implementation of Umpire and present case studies from the integration of Umpire into applications that are currently running on Sierra.

97 MATHEMATICS AND COMPUTING↗

Programmable simulations of molecules and materials with reconfigurable quantum processors

Simulations of quantum chemistry and quantum materials are believed to be among the most important applications of quantum information processors. However, realizing practical quantum advantage for such problems is challenging because of the prohibitive computational cost of programming typical problems into quantum hardware. Here we introduce a simulation framework for strongly correlated quantum systems represented by model spin Hamiltonians that uses reconfigurable qubit architectures to simulate real-time dynamics in a programmable way. Our approach also introduces an algorithm for extracting chemically relevant spectral properties via classical co-processing of quantum measurement results. We develop a digital–analogue simulation toolbox for efficient Hamiltonian time evolution using digital Floquet engineering and hardware-optimized multi-qubit operations to accurately realize complex spin–spin interactions. As an example, we propose an implementation based on Rydberg atom arrays. In addition, we show how detailed spectral information can be extracted from the dynamics through snapshot measurements and single-ancilla control, enabling the evaluation of excitation energies and finite-temperature susceptibilities from a single dataset. To illustrate the approach, we show how to use the method to compute key properties of a polynuclear transition-metal catalyst and two-dimensional magnetic materials.

74 ATOMIC AND MOLECULAR PHYSICS↗

IPAD products and implications for the future

The betterment of productivity through the improvement of product quality and the reduction of cost is addressed. Productivity improvement is sought through (1) reduction of required resources, (2) improved ask results through the management of such saved resources, (3) reduced downstream costs through manufacturing-oriented engineering, and (4) lowered risks in the making of product design decisions. The IPAD products are both hardware architecture and software distributed over a number of heterogeneous computers in this architecture. These IPAD products are described in terms of capability and engineering usefulness. The future implications of state-of-the-art IPAD hardware and software architectures are discussed in terms of their impact on the functions and on structures of organizations concerned with creating products.

Miller, R. E., Jr.↗

Low Size, Weight, and Power Neuromorphic Computing to Improve Combustion Engine Efficiency

Neuromorphic computing offers one path forward for AI at the edge. However, accessing and effectively utilizing a neuromorphic hardware platform is non-trivial. In this work, we present a complete pipeline for neuromorphic computing at the edge, including a small, inexpensive, low-power, FPGA-based neuromorphic hardware platform, a training algorithm for designing spiking neural networks for neuromorphic hardware, and a software framework for connecting those components. We demonstrate this pipeline on a real-world application, engine control for a spark-ignition internal combustion engine. We illustrate how we connect engine simulations with neuromorphic hardware simulations and training software to produce hardware-compatible spiking neural networks that perform engine control to improve fuel efficiency. We present initial results on the performance of these spiking neural networks and illustrate that they outperform open-loop engine control. We also give size, weight, and power estimates for a deployed solution of this type.

Schuman, Catherine↗