Search NASA⌕ Search

SEARCH · Search NASA

Results for “machine architecture”

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 397 records · Page 22
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Evaluating Spatial Accelerator Architectures with Tiled Matrix-Matrix Multiplication.

There is a growing interest in custom spatial accelerators for machine learning applications. These accelerators employ a spatial array of processing elements (PEs) interacting via custom buffer hierarchies and networks-on-chip. The efficiency of these accelerators comes from employing optimized dataflow (i.e., spatial/temporal partitioning of data across the PEs and fine-grained scheduling) strategies to optimize data reuse. The focus of this work is to evaluate these accelerator architectures using a tiled general matrix-matrix multiplication (GEMM) kernel. To do so, we develop a framework that finds optimized mappings (dataflow and tile sizes) for a tiled GEMM for a given spatial accelerator and workload combination, leveraging an analytical cost model for runtime and energy. Finally, our evaluations over five spatial accelerators demonstrate that the tiled GEMM mappings systematically generated by our framework achieve high performance on various GEMM workloads and accelerators.

42 ENGINEERING↗
DOE OSTI2021

Evaluating Spatial Accelerator Architectures with Tiled Matrix-Matrix Multiplication

There is a growing interest in custom spatial accelerators for machine learning applications. These accelerators employ a spatial array of processing elements (PEs) interacting via custom buffer hierarchies and networks-on-chip. The efficiency of these accelerators comes from employing optimized dataflow (i.e., spatial/temporal partitioning of data across the PEs and fine-grained scheduling) strategies to optimize data reuse. The focus of this work is to evaluate these accelerator architectures using a tiled general matrix-matrix multiplication (GEMM) kernel. To do so, we develop a framework that finds optimized mappings (dataflow and tile sizes) for a tiled GEMM for a given spatial accelerator and workload combination, leveraging an analytical cost model for runtime and energy. Our evaluations over five spatial accelerators demonstrate that the tiled GEMM mappings systematically generated by our framework achieve high performance on various GEMM workloads and accelerators.

43 PARTICLE ACCELERATORS↗
DOE OSTI2023

Highly-scalable GPU-accelerated compressible reacting flow solver for modeling high-speed flows

Emerging supercomputing systems utilize a combination of central processing units (CPUs) and graphics processing units (GPUs) in an effort to reach exascale capabilities while minimizing the energy footprint of operating such systems. Such heterogeneous machines introduce new challenges for fluids solvers because the hardware architecture and operation of a GPU are fundamentally different from conventional CPUs. In this work, a general approach for efficient implementation of finite-volume based reacting flow solvers on such heterogeneous systems is presented. Three main challenges, namely, data access pattern, thread divergence, and thread safety, are addressed. Since compressible reacting flows require special methods to deal with chemical reactions, hyperbolic and nonlinear convection terms, and the presence of turbulence, specific algorithms that ensure GPU-based efficiency are developed. The approach is demonstrated on the widely available OpenFOAM open source software by modifying core algorithms for GPU accessibility. The scalability of the resulting solver, is demonstrated using practical test cases, including flow through a scramjet engine and the dynamics of a rotating detonation engine. Here, the solver provides near-ideal scaleup on a large number of GPUs (>3000), and extremely efficient use of the GPUs, with throughput nearly a constant even when processing a large number of control volumes.

42 ENGINEERING↗
DOE OSTI2022

Development and performance of a HemeLB GPU code for human-scale blood flow simulation

In recent years, it has become increasingly common for high performance computers (HPC) to possess some level of heterogeneous architecture - typically in the form of GPU accelerators. In some machines these are isolated within a dedicated partition, whilst in others they are integral to all compute nodes - often with multiple GPUs per node - and provide the majority of a machine's compute performance. In light of this trend, it is becoming essential that codes deployed on HPC are updated to execute on accelerator hardware. Here, in this paper, we introduce a GPU implementation of the 3D blood flow simulation code HemeLB that has been developed using CUDA C++. We demonstrate how taking advantage of NVIDIA GPU hardware can achieve significant performance improvements compared to the equivalent CPU only code on which it has been built whilst retaining the excellent strong scaling characteristics that have been repeatedly demonstrated by the CPU version. With HPC positioned on the brink of the exascale era, we use HemeLB as a motivation to provide a discussion on some of the challenges that many users will face when deploying their own applications on upcoming exascale machines.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Quantifying uncertainty for deep learning based forecasting and flow-reconstruction using neural architecture search ensembles

Classical problems in computational physics such as data-driven forecasting and signal reconstruction from sparse sensors have recently seen an explosion in deep neural network (DNN) based algorithmic approaches. However, most DNN models do not provide uncertainty estimates, which are crucial for establishing the trustworthiness of these techniques in downstream decision making tasks and scenarios. In recent years, ensemble-based methods have achieved significant success for the uncertainty quantification in DNNs on a number of benchmark problems. However, their performance on real-world applications remains under-explored. In this work, we present an automated approach to DNN discovery and demonstrate how this may also be utilized for ensemble-based uncertainty quantification. Specifically, we propose the use of a scalable neural and hyperparameter architecture search for discovering an ensemble of DNN models for complex dynamical systems. We highlight how the proposed method not only discovers high-performing neural network ensembles for our tasks, but also quantifies uncertainty seamlessly. This is achieved by using genetic algorithms and Bayesian optimization for sampling the search space of neural network architectures and hyperparameters. Subsequently, a model selection approach is used to identify candidate models for an ensemble set construction. Afterwards, a variance decomposition approach is used to estimate the uncertainty of the predictions from the ensemble. We demonstrate the feasibility of this framework for two tasks — forecasting from historical data and flow reconstruction from sparse sensors for the sea-surface temperature. In conclusion, we demonstrate superior performance from the ensemble in contrast with individual high-performing models and other benchmarks.

Deep ensembles↗
DOE OSTI2025

The design space of E(3)-equivariant atom-centred interatomic potentials

Abstract Molecular dynamics simulation is an important tool in computational materials science and chemistry, and in the past decade it has been revolutionized by machine learning. This rapid progress in machine learning interatomic potentials has produced a number of new architectures in just the past few years. Particularly notable among these are the atomic cluster expansion, which unified many of the earlier ideas around atom-density-based descriptors, and Neural Equivariant Interatomic Potentials (NequIP), a message-passing neural network with equivariant features that exhibited state-of-the-art accuracy at the time. Here we construct a mathematical framework that unifies these models: atomic cluster expansion is extended and recast as one layer of a multi-layer architecture, while the linearized version of NequIP is understood as a particular sparsification of a much larger polynomial model. Our framework also provides a practical tool for systematically probing different choices in this unified design space. An ablation study of NequIP, via a set of experiments looking at in- and out-of-domain accuracy and smooth extrapolation very far from the training data, sheds some light on which design choices are critical to achieving high accuracy. A much-simplified version of NequIP, which we call BOTnet (for body-ordered tensor network), has an interpretable architecture and maintains its accuracy on benchmark datasets.

Computer Science↗
DOE OSTI2021

Elucidating the local atomic and electronic structure of amorphous oxidized superconducting niobium films

Qubits made from superconducting materials are a mature platform for quantum information science application, such as quantum computing. However, material-based losses are now a limiting factor in reaching the coherence times needed for applications. In particular, knowledge of the atomistic structure and properties of the circuit materials is needed to identify, understand, and mitigate material-based decoherence channels. Here, we characterize the atomic structure of the native oxide film formed on Nb resonators by comparing fluctuation electron microscopy experiments to density functional theory calculations, finding that an amorphous layer is consistent with an Nb 2 O 5 stoichiometry. Comparing x-ray absorption measurements at the Oxygen K edge with first-principles calculations, we find evidence of d-type magnetic impurities in our sample, known to cause impedance in proximal superconductors. This work identifies the structural and chemical composition of the oxide layer grown on Nb superconductors and shows that soft x-ray absorption can fingerprint magnetic impurities in these superconducting systems.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2020

Phase I Final Technical Report on Energy-Efficient Reconfigurable Universal Accelerator Interconnect

It is well known that application specific computing systems, optimally designed and configured for a given workload, offer much higher energy-efficiency and throughput than general purpose systems. In modern computing systems, heterogeneous computing systems have emerged that exploit the energy and performance benefits of combining various different domain-specific processor architectures. Application domains such as high-performance computing and machine learning now process terabyte-sized data sets, requiring enormous processing and memory resources. These applications have very high-power consumption due to bottlenecks in the electrical interconnection between processing units. This project aims to reduce both communication energy and latency by integrating universally available accelerators with silicon photonics. It also aims to increase system throughput by exploiting emerging technologies in silicon photonic reconfigurable interconnects; this will allow the system to balance itself in real time to accommodate changes in workloads and data flows.

2.5D/3D integration↗
NASA NTRS1988

Systems autonomy

Information on systems autonomy is given in viewgraph form. Information is given on space systems integration, intelligent autonomous systems, automated systems for in-flight mission operations, the Systems Autonomy Demonstration Project on the Space Station Thermal Control System, the architecture of an autonomous intelligent system, artificial intelligence research issues, machine learning, and real-time image processing.

Lum, Henry, Jr.↗
NASA NTRS1988

The integrated analysis capability (IAC Level 2.0)

The critical data management issues involved in the development of the integral analysis capability (IAC), Level 2, to support the design analysis and performance evaluation of large space structures, are examined. In particular, attention is given to the advantages and disadvantages of the formalized data base; merging of the matrix and relational data concepts; data types, query operators, and data handling; sequential versus direct-access files; local versus global data access; programming languages and host machines; and data flow techniques. The discussion also covers system architecture, recent system level enhancements, executive/user interface capabilities, and technology applications.

Frisch, Harold P.↗
NASA NTRS1991

Human factors issues for interstellar spacecraft

Developments in research on space human factors are reviewed in the context of a self-sustaining interstellar spacecraft based on the notion of traveling space settlements. Assumptions about interstellar travel are set forth addressing costs, mission durations, and the need for multigenerational space colonies. The model of human motivation by Maslow (1970) is examined and directly related to the design of space habitat architecture. Human-factors technology issues encompass the human-machine interface, crew selection and training, and the development of spaceship infrastructure during transtellar flight. A scenario for feasible instellar travel is based on a speed of 0.5c, a timeframe of about 100 yr, and an expandable multigenerational crew of about 100 members. Crew training is identified as a critical human-factors issue requiring the development of perceptual and cognitive aids such as expert systems and virtual reality.

Cohen, Marc M.↗
NASA NTRS1992

Multidisciplinary computational aerosciences

As the challenges of single disciplinary computational physics are met, such as computational fluid dynamics, computational structural mechanics, computational propulsion, computational aeroacoustics, computational electromagnetics, etc., scientists have begun investigating the combination of these single disciplines into what is being called multidisciplinary computational aerosciences (MCAS). The combination of several disciplines not only offers simulation realism but also formidable computational challenges. The solution of such problems will require computers orders of magnitude larger than those currently available. Such computer power can only be supplied by massively parallel machines because of the current speed-of-light limitation of conventional serial systems. Even with such machines, MCAS problems will require hundreds of hours for their solution. To efficiently utilize such a machine, research is required in three areas that include parallel architectures, systems software, and applications software. The main emphasis of this paper is the applications software element. Examples that demonstrate application software for multidisciplinary problems currently being solved at NASA Ames Research Center are presented. Pacing items for MCAS are discussed such as solution methodology, physical modeling, computer power, and multidisciplinary validation experiments.

Kutler, Paul↗
NASA NTRS1991

Fault tolerant sequential circuits using sequence invariant state machines

The idea of introducing redundancy to improve the reliability of digital systems originates from papers published in the 1950's. Since then, redundancy has been recognized as a realistic means for constructing reliable systems. A method using redundancy to reconfigure the Sequency Invariant State Machine (SISM) to achieve fault tolerance is introduced. This new architecture is most useful in space applications, where recovery rather than replacement of faulty modules is the only means of maintenance.

Alahmad, M.↗
NASA NTRS1994

NASA STI Program Coordinating Council Twelfth Meeting: Standards

The theme of this NASA Scientific and Technical Information Program Coordinating Council Meeting was standards and their formation and application. Topics covered included scientific and technical information architecture, the Open Systems Interconnection Transmission Control Protocol/Internet Protocol, Machine-Readable Cataloging (MARC) open system environment procurement, and the Government Information Locator Service.

Source record↗
NASA NTRS1996

Basic Requirements for Systems Software Research and Development

Our success over the past ten years evaluating and developing advanced computing technologies has been due to a simple research and development (R/D) model. Our model has three phases: (a) evaluating the state-of-the-art, (b) identifying problems and creating innovations, and (c) developing solutions, improving the state- of-the-art. This cycle has four basic requirements: a large production testbed with real users, a diverse collection of state-of-the-art hardware, facilities for evalua- tion of emerging technologies and development of innovations, and control over system management on these testbeds. Future research will be irrelevant and future products will not work if any of these requirements is eliminated. In order to retain our effectiveness, the numerical aerospace simulator (NAS) must replace out-of-date production testbeds in as timely a fashion as possible, and cannot afford to ignore innovative designs such as new distributed shared memory machines, clustered commodity-based computers, and multi-threaded architectures.

Kuszmaul, Chris↗
NASA NTRS2020

Control Demonstration of Multiple Doubly Fed Induction Motors for Hybrid Electric Propulsion

The Convergent Aeronautics Solutions (CAS) High Voltage Hybrid Electric Propulsion (HVHEP) activity was formulated to support the move into future hybrid-electric aircraft. The goal of this high-voltage activity is to develop a new alternating current (AC) power architecture to support the needs of higher efficiency and lower emissions aircraft. This proposed architecture will adopt the use of the doubly fed induction machine (DFIM) for propulsor drive motor application. The DFIMs are attractive for several reasons, including but not limited to, the ability to self-start, the ability to operate subsynchronously and supersynchronously, and the requirement for power converters rated at a fraction of those required in a corresponding direct current system, dependent on the range of operation. The focus of this report is based specifically on the presentation and analysis of a novel strategy, which allows for independent operation of multiple doubly fed induction motors from a common AC bus

David J Sadey↗
DOE OSTI2024

Forensic Analysis of SOHO Router Binaries

Small Office/Home Office (SOHO) routers are used by millions of consumers across the United States, and are commensurately vulnerable. Forensic analysis of SOHO router firmware helps to understand and mitigate those vulnerabilities. This poster focused particularly on analysis of BusyBox executables, a software suite that provides several Unix utilities in a single file. Three main tools were used to analyze the binaries. BinWalk was used to extract the files, but also to build entropy graphs, extract Linux kernel images, and identify CPU architectures; WiiBin processed the binaries to find endianness, architecture, the percent compressed/encrypted, and compiler data; and @DisCo, a machine learning tool used to determine function similarity in disassembled binaries, analyzed similarities and determined versions of extracted BusyBox files from each router. These tools found that venders from all five routers utilized the same version of the BusyBox software across different firmware updates, demonstrating the importance of constant firmware scrutiny to protect against security vulnerabilities.

24 POWER TRANSMISSION AND DISTRIBUTION↗