Search NASA⌕ Search

SEARCH · Search NASA

Results for “memory bottleneck”

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 55 records · Page 3
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Energy-efficient Mott activation neuron for full-hardware implementation of neural networks

To circumvent the von Neumann bottleneck, substantial progress has been made towards in-memory computing with synaptic devices. However, compact nanodevices implementing non-linear activation functions are required for efficient full-hardware implementation of deep neural networks. Here, in this work, we present an energy-efficient and compact Mott activation neuron based on vanadium dioxide and its successful integration with a conductive bridge random access memory (CBRAM) crossbar array in hardware. The Mott activation neuron implements the rectified linear unit function in the analogue domain. The neuron devices consume substantially less energy and occupy two orders of magnitude smaller area than those of analogue complementary metal–oxide semiconductor implementations. The LeNet-5 network with Mott activation neurons achieves 98.38% accuracy on the MNIST dataset, close to the ideal software accuracy. We perform large-scale image edge detection using the Mott activation neurons integrated with a CBRAM crossbar array. Our findings provide a solution towards large-scale, highly parallel and energy-efficient in-memory computing systems for neural networks.

electrical and electronic engineering↗
DOE OSTI2020

Efficient Data Management in Neutron Scattering Data Reduction Workflows at ORNL

Oak Ridge National Laboratory (ORNL) experimental neutron science facilities produce 1.2 TB a day of raw event-based data that is stored using the standard metadata-rich NeXus schema built on top of the HDF5 file format. Performance of several data reduction workflows is largely determined by the amount of time spent on the loading and processing algorithms in Mantid, an open-source data analysis framework used across several neutron sciences facilities around the world. The present work introduces new data management algorithms to address identified input output (I/O) bottlenecks on Mantid. First, we introduce an in-memory binary-tree metadata index that resemble NeXus data access patterns to provide a scalable search and extraction mechanism. Second, data encapsulation in Mantid algorithms is optimally redesigned to reduce the total compute and memory runtime footprint associated with metadata I/O reconstruction tasks. Results from this work show speed ups in wall-clock time on ORNL data reduction workflows, ranging from 11% to 30% depending on the complexity of the targeted instrument-specific data. Nevertheless, we highlight the need for more research to address reduction challenges as experimental data volumes increase.

Godoy, William↗
DOE OSTI2024

Developing And Scaling an OpenFOAM Model to Study Turbulent Flow in a HFIR Coolant Channel

Improving the understanding of how computational fluid dynamics (CFD) direct numerical simulations (DNS) of flows in the High Flux Isotope Reactor (HFIR) perform when run in parallel using the high performance computing (HPC) platform Summit at the Oak Ridge Leadership Computing Facility (OLCF) is of particular importance to boost the computational tools used to support HFIR conversion to low enriched fuel (LEU). Evaluation of scaling performance was driven by the increasing importance of graphics processing unit (GPU) usage in HPC, which is becoming the standard for modern supercomputers such as Summit. The desired results are to obtain a strong positive correlation between the computational resources dedicated to a problem and the relative speed-up of the simulation in comparison to a benchmark. This capability will allow substantially improvement in HFIR flow analytical capabilities, specifically when predicting turbulence properties at high Reynolds numbers. The study leverages previous simulation results performed with code PHASTA (finite element) on HPC platforms Cori (NERSC) and Theta (ALCF) [1] with computing options provided in the computing platform OpenFOAM (finite volume) at OLCF. Transitioning from PHASTA to OpenFOAM will (1) eliminate dependence on third-party software for mesh generation and manipulation, (2) reduce resource needs by employing modern architectures, and (3) build expertise for future modeling of HFIR-specific problems like heat transfer in involute geometry, entrance effects, flow structure in channel corners, and so on—all important issues when defining the available thermal margins in the transition to LEU. CPUs and GPUs differ significantly in their architecture and utilization, as discussed in the literature [2]. The most important differences are in the approach to computations and their memory. A single GPU contains a large quantity of cores, enabling it to perform with a much higher throughput than a CPU, but execution requires a different approach. GPU codes execute instructions using the Single-Instruction Multiple-Thread (SIMT) approach in which a single instruction is used for groups of threads called warps. A warp typically consists of 32 threads which must execute the same set of instructions, although on separate threads. Alternately, a CPU has far fewer cores that are much more flexible in their operation, excelling at quickly performing more complex serial computations. This is why GPUs have greater throughput when properly utilized. The second important difference is seen when comparing their memory spaces. Limited memory allocations and CPU–GPU communications cause a significant bottleneck in GPU-accelerated programs. Further study was required to properly take advantage of GPU resources. A comprehensive analysis of code performance and the model-specific features of turbulence constitutes the core of this work. In this study, a DNS simulation of HFIR channel turbulence was performed with the finite volume CFD code OpenFOAM v2112 and CUDA v11.0 on Red Hat Enterprise Linux v8.2. The OpenFOAM installation had AMGx integrated to enable GPU acceleration and utilizes the PETSc4FOAM library. The computational resources and the problem size were scaled on CPU and CPU + GPU architectures to gain a better understanding of the performance of a DNS problem on modern computing hardware. The study aimed to analyze the scaling of the code exclusively on CPUs and then to examine the scaling of the codes with GPU acceleration enabled. Scaling studies included CPU and GPU acceleration on a mesh of varying resolution to analyze the impact of problem size relative to computational resources. In the course of preparing the GPU configuration on Summit, mainly using the AMGX solvers, difficulties were encountered stemming from constant changes resulting from extensive ongoing development activities and the changing environment. This resulted in the inability to complete the GPU portion of the work. The code was compiled and tested, but production runs to assess acceleration were not performed because the used discretional compute time allocation expired as year-end approached. The Summit HPC platform is scheduled for decommissioning in 2024, making it unattractive for future use with Nvidia-based GPUs. Therefore, the work will be moved onto NERSC machines in FY24. An application was prepared and submitted, and sufficient node-hours were awarded to continue the research in the next calendar year. This report summarizes work performed thus far, which mostly focused on CPU OpenFOAM computing.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2026

Ferroelectric Fractals: Switching Mechanism of Wurtzite AlN

The advent of wurtzite ferroelectrics is enabling new ferroelectric devices for computer memory that have the potential to bypass the von Neumann bottleneck due to their robust polarization and silicon compatibility. However, the atomistic switching mechanism of wurtzites is still undetermined due to the limitations of density functional theory simulation size and experimental temporal and spatial resolution. Thus, physics-informed materials engineering to reduce coercive field and breakdown in these devices has been limited. In this work, the atomistic mechanism of domain wall migration and domain growth in aluminum nitride-based wurtzites is uncovered using molecular dynamics and Monte Carlo simulations. We reveal the anomalous switching mechanism of fast 1D single columns of atoms propagating from a slow-moving 2D fractallike domain wall. We find that the critical nucleus is a single aluminum ion that breaks its bond with one nitrogen and bonds to another nitrogen; this creates a cascade that flips atoms directly only in the same column, due to the extreme locality (sharpness) of the domain walls in wurtzites. We further show how the fractallike shape of the domain wall in the 2D plane breaks assumptions in the Kolmogorov, Avrami, and Ishibashi (KAI) model and leads to the anomalously fast switching in wurtzite structured ferroelectrics.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Two new SciDAC institutes promote mathematical tools and software technology for high-performance computing

Bigger is often said to be better, and the newest extreme-scale computers certainly are bigger, with millions of processing units. Moreover, the breadth of science performed on the U.S. Department of Energy (DOE) computing facilities is expanding, with new technology such as artificial intelligence emerging. These advances are exciting, creating new opportunities for scientific discovery; however, they also raise new questions for scientists who want to exploit these advances for tackling more complex problems. Will my simulation code be able to utilize the accelerators in extreme-scale computing systems? Can I take advantage of the deepening memory hierarchy in heterogeneous processors? Is there a way around bottlenecks caused by the widening ratio of peak floating-point operations per second to I/0 bandwidth? How can I manage my huge amounts of data effectively? Can I analyze data in situ, or must I transfer it to offline storage for later analysis? To address such questions, DOE announced that it is providing $57.5 million over the next five years for two multidisciplinary teams — FASTMath and RAPIDS2 — to develop new tools and techniques to harness supercomputers for scientific discovery. The teams, called SciDAC Institutes, are part of the Scientific Discovery through Advanced Computing program.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1986

Application of data flow concepts to a multigrid solver for the Euler equations

In this study a multigrid solver for Euler equations (FLO52R) was examined to determine its performance potential on a hypothetical computer using a data flow architecture. The proposed computer would require massive parallelism to realize its design performance. On the other hand this parallelism would be more easily realized than with a conventional vector processor such as the Cray-1S. Several changes to the proposed design substantially alleviated most of the remaining bottlenecks to parallel processing. Other changes allowed clearer definition of memory access and disk I/O. Finally, a portion of the algorithm was rewritten to improve parallel performance. With these changes, performance levels approaching that of a Cray-1S may be possible for a computer costing far less. Estimates are given for overall speed, memory, and network bandwidth, and for instruction memory requirements.

Merriam, M. L.↗
NASA NTRS1990

Performance measurements of the first RAID prototype

The performance is examined of Redundant Arrays of Inexpensive Disks (RAID) the First, a prototype disk array. A hierarchy of bottlenecks was discovered in the system that limit overall performance. The most serious is the memory system contention on the Sun 4/280 host CPU, which limits array bandwidth to 2.3 MBytes/sec. The array performs more successfully on small random operations, achieving nearly 300 I/Os per second before the Sun 4/280 becomes CPU limited. Other bottlenecks in the system are the VME backplane, bandwidth on the disk controller, and overheads associated with the SCSI protocol. All are examined in detail. The main conclusion is that to achieve the potential bandwidth of arrays, more powerful CPU's alone will not suffice. Just as important are adequate host memory bandwidth and support for high bandwidth on disk controllers. Current disk controllers are more often designed to achieve large numbers of small random operations, rather than high bandwidth. Operating systems also need to change to support high bandwidth from disk arrays. In particular, they should transfer data in larger blocks, and should support asynchronous I/O to improve sequential write performance.

Chervenak, Ann L.↗
DOE OSTI2022

On-Surface Single-Molecule Identification of Mass-Selected Cyclodextrin-Supported Polyoxovanadates for Multistate Resistive-Switching Memory Applications

Polyoxovanadate anions immobilized on conducting substrates are promising candidates for molecule-based memory technologies. However, strong electronic interactions with the substrate and the formation of larger agglomerates remain bottlenecks on the way to their technical application. We demonstrate that ion soft-landing of mass selected cyclodextrin- functionalized hexavanadates enables to deposit these host-guest complexes as single memory units on surfaces without their quaternary ammonium countercations from solution. The cyclodextrin provides a stabilizing “shell” for the dianions. Further, multistate switching evidenced by the change in molecular resistance has been detected using scanning tunneling microscopy and spectroscopy and attributed to reduction of individual vanadium (+V) centers.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Reducing Memory Consumption in Calico with Shared Memory

This document details the work to reduce memory consumption in Calico. Calico is SimTools’ Constructive Solid Geometry (CSG) and geometry painting library. It is primarily used to paint material volume fractions in the Eulerian meshes of the physics codes. Calico provides point-in-body checks for the geometry supplied by an Oso model, which are then aggregated by the host codes. In addition, Calico can be used to build Oso models and is used by Ingen for that purpose. Oso models, and thus Calico, provide support for various CSG primitives such as spheres, cylinders, surfaces generated by rotating tabular curve data, and STL files as well as binary combinations of those primitives. Prior to refactoring Calico will run out of memory on CTS-1 machines when 36 MPI ranks are used per node when reading STL models on the order of 1.5 GB. This limitation is a bottleneck in designer workflow. This problem has been alleviated through the use of data structures to both reduce memory consumption and to leverage MPI-3 shared memory. This report details the data structures targeted for refactoring in Calico, the methods and implementation details for reducing memory consumption and leveraging shared memory, and results for one test problem. Results show a memory reduction when loading a 1 GB STL file by a factor of 27.5, from 93.4 to 3.4 GB.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Hutchinson Trace Estimation for high-dimensional and high-order Physics-Informed Neural Networks

Physics-Informed Neural Networks (PINNs) have proven effective in solving partial differential equations (PDEs), especially when some data are available by seamlessly blending data and physics. However, extending PINNs to high-dimensional and even high-order PDEs encounters significant challenges due to the computational cost associated with automatic differentiation in the residual loss function calculation. Herein, we address the limitations of PINNs in handling high-dimensional and high-order PDEs by introducing the Hutchinson Trace Estimation (HTE) method. Starting with the second-order high-dimensional PDEs, which are ubiquitous in scientific computing, HTE is applied to transform the calculation of the entire Hessian matrix into a Hessian vector product (HVP). This approach not only alleviates the computational bottleneck via Taylor-mode automatic differentiation but also significantly reduces memory consumption from the Hessian matrix to an HVP’s scalar output. We further showcase HTE’s convergence to the original PINN loss and its unbiased behavior under specific conditions. Comparisons with the Stochastic Dimension Gradient Descent (SDGD) highlight the distinct advantages of HTE, particularly in scenarios with significant variability and variance among dimensions. We further extend the application of HTE to higher-order and higher-dimensional PDEs, specifically addressing the biharmonic equation. By employing tensor-vector products (TVP), HTE efficiently computes the colossal tensor associated with the fourth-order high-dimensional biharmonic equation, saving memory and enabling rapid computation. The effectiveness of HTE is illustrated through experimental setups, demonstrating comparable convergence rates with SDGD under memory and speed constraints. Additionally, HTE proves valuable in accelerating the Gradient-Enhanced PINN (gPINN) version as well as the Biharmonic equation. Overall, HTE opens up a new capability in scientific machine learning for tackling high-order and high-dimensional PDEs.

Curse of dimensionality↗
DOE OSTI2020

Pass-efficient methods for compression of high-dimensional turbulent flow data

The future of high-performance computing, specifically on future Exascale computers, will presumably see memory capacity and bandwidth fail to keep pace with data generated, for instance, from massively parallel partial differential equation (PDE) systems. Current strategies proposed to address this bottleneck entail the omission of large fractions of data, as well as the incorporation of in situ compression algorithms to avoid overuse of memory. To ensure that post-processing operations are successful, this must be done in a way that a sufficiently accurate representation of the solution is stored. Moreover, in situations where the input/output system becomes a bottleneck in analysis, visualization, etc., or the execution of the PDE solver is expensive, the number of passes made over the data must be minimized. In the interest of addressing this problem, this work focuses on the utility of pass-efficient, parallelizable, low-rank, matrix decomposition methods in compressing high-dimensional simulation data from turbulent flows. Additionally, a particular emphasis is placed on using coarse representation of the data – compatible with the PDE discretization grid – to accelerate the construction of the low-rank factorization. This includes the presentation of a novel single-pass matrix decomposition algorithm for computing the so-called interpolative decomposition. The methods are described extensively and numerical experiments on two turbulent channel flow data are performed. In the first (unladen) channel flow case, compression factors exceeding 400 are achieved while maintaining accuracy with respect to first- and second-order flow statistics. In the particle-laden case, compression factors of 100 are achieved and the compressed data is used to recover particle velocities. These results show that these compression methods can enable efficient computation of various quantities of interest in both the carrier and disperse phases.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2024

Optimal control of large quantum systems: assessing memory and runtime performance of GRAPE

Abstract Gradient Ascent Pulse Engineering (GRAPE) is a popular technique in quantum optimal control, and can be combined with automatic differentiation (AD) to facilitate on-the-fly evaluation of cost-function gradients. We illustrate that the convenience of AD comes at a significant memory cost due to the cumulative storage of a large number of states and propagators. For quantum systems of increasing Hilbert space size, this imposes a significant bottleneck. We revisit the strategy of hard-coding gradients in a scheme that fully avoids propagator storage and significantly reduces memory requirements. Separately, we present improvements to numerical state propagation to enhance runtime performance. We benchmark runtime and memory usage and compare this approach to AD-based implementations, with a focus on pushing towards larger Hilbert space sizes. The results confirm that the AD-free approach facilitates the application of optimal control for large quantum systems which would otherwise be difficult to tackle.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
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 NTRS1993

Interfacing a high performance disk array file server to a Gigabit LAN

Our previous prototype, RAID-1, identified several bottlenecks in typical file server architectures. The most important bottleneck was the lack of a high-bandwidth path between disk, memory, and the network. Workstation servers, such as the Sun-4/280, have very slow access to peripherals on busses far from the CPU. For the RAID-2 system, we addressed this problem by designing a crossbar interconnect, Xbus board, that provides a 40MB/s path between disk, memory, and the network interfaces. However, this interconnect does not provide the system CPU with low latency access to control the various interfaces. To provide a high data rate to clients on the network, we were forced to carefully and efficiently design the network software. A block diagram of the system hardware architecture is given. In the following subsections, we describe pieces of the RAID-2 file server hardware that had a significant impact on the design of the network interface.

Seshan, Srinivasan↗
DOE OSTI2025

High-fidelity remote entanglement of trapped atoms mediated by time-bin photons

Abstract Photonic interconnects between quantum processing nodes are likely the only way to achieve large-scale quantum computers and networks. The bottleneck in such an architecture is the interface between well-isolated quantum memories and flying photons. We establish high-fidelity entanglement between remotely separated trapped atomic qubit memories, mediated by photonic qubits stored in the timing of their pulses. Such time-bin encoding removes sensitivity to polarization errors, enables long-distance quantum communication, and is extensible to quantum memories with more than two states. Using a measurement-based error detection process and suppressing a fundamental source of error due to atomic recoil, we achieve an entanglement fidelity of 97% and show that fundamental limits due to atomic recoil still allow fidelities in excess of 99.9%.

Science & Technology - Other Topics↗
DOE OSTI2020

Population Count on Intel® CPU, GPU, and FPGA

Population count is a primitive used in many applications. Commodity processors have dedicated instructions for achieving high-performance population count. Motivated by the productivity of high-level synthesis and the importance of population count, in this paper we investigated the OpenCL implementations of population count algorithms, and evaluated their performance and resource utilizations on an FPGA. Based on the results, we select the most efficient implementation. Then we derived a reduction pattern from a representative application of population count. We parallelized the reduction with atomic functions, and optimized it with vectorized memory accesses, tree reduction, and compute-unit duplication. We evaluated the performance of the reduction kernel on an Intel (R) Xeon (R) CPU and an Intel (R) IrisT Pro integrated GPU, and an FPGA card that features an Intel (R) Arria (R) 10 FPGA. When DRAM memory bandwidth is comparable on the three computing platforms, the FPGA can achieve the highest kernel performance for large workload. On the other hand, we described performance bottlenecks on the FPGA. To make FPGAs more competitive in raw performance compared to high-performant CPU and GPU platforms, it is important to increase external memory bandwidth, minimize data movement between a host and a device, and reduce OpenCL runtime overhead on an FPGA.

heterogeneous computing↗
DOE OSTI2025

Anodized Aluminum Oxide Membrane Ionic Memristors

Memory effect in ion transport (IT) at the solid–solution interface is uniquely attractive in that the conductance depends on or “memorizes” the previous states. Hysteretic and rectified transport properties offer exciting potential to developing advanced iontronics and neuromorphic functions, improving the efficiency of energy conversion and electrochemical processes, and overcoming the selectivity-throughput bottleneck in the enrichment of low abundant species for environment- and energy-friendly separations, among others. Herein, memory effects are discovered in the rectified electrokinetic IT through anodized aluminum oxide (AAO) membranes containing densely packed highly ordered nanochannels (10 10 per cm 2 ). Characteristic memristor responses of pinched current–potential loops are resolved in voltammetric experiments and successfully reproduced through finite element simulation. Excitatory and inhibitory conductance states are shown to arise from the enrichment and depletion of mobile charge carriers. Structurewise, the transport symmetry is broken by the barrier oxide layer (BOL) on the one end of the cylindrical nanochannels across the AAO membranes. Charge selectivity is attributed to the gradient(s) of the space charge density across the BOL characterized by depth profiling via X-ray photoelectron spectroscopy analysis. The space charge gradient(s) overcomes the fundamental limitation of widely exploited surface charge effects to enable intense rectification and hysteresis prevailing at very high ionic concentrations up to 1–2 M. A new strategy is developed for controlling the preferential IT direction and selectivity via counterion intercalation and extraction/exchange. Mechanistic understanding is further confirmed through parameter variations such as potential scan rate and ionic strength, which also demonstrates convenient controls of the related functions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
NASA NTRS1992

Memory Network For Distributed Data Processors

Universal Memory Network (UMN) is modular, digital data-communication system enabling computers with differing bus architectures to share 32-bit-wide data between locations up to 3 km apart with less than one millisecond of latency. Makes it possible to design sophisticated real-time and near-real-time data-processing systems without data-transfer "bottlenecks". This enterprise network permits transmission of volume of data equivalent to an encyclopedia each second. Facilities benefiting from Universal Memory Network include telemetry stations, simulation facilities, power-plants, and large laboratories or any facility sharing very large volumes of data. Main hub of UMN is reflection center including smaller hubs called Shared Memory Interfaces.

Bolen, David↗