Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “gpu”

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 1,063 records · Page 59
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Producing High-fidelity Synthetic Population Ensembles at Scale

Used within social simulations, synthetic population ensembles enable uncertainty quantification (UQ) methods for obtaining more robust model inference and prediction. A synthetic population ensemble is a series of plausible virtual reconstructions of an area’s population at the granularity of people and residences, generated stochastically to preserve privacy of the source population survey’s respondents. In this paper, we demonstrate the production of large synthetic population ensembles for the US via Oak Ridge National Laboratory’s UrbanPop framework to support modeling of high spatial resolution energy affordability metrics from nationwide social surveys in collaboration with the fusionACS project. Our initial task involves creating ensembles for 17 US metropolitan areas, each consisting of 41 population instances (a base realization and 40 replicates). To accomplish this task at scale, we configured an integrated system comprised of a research cloud, virtual containerization, GPU-enhanced functionality, and a dual API/CLI to interact with UrbanPop’s maturing Likeness Python ecosystem. We observe a reduction in theoretical execution time while maintaining high-fidelity approximations of residential totals by metropolitan area and the demographic characteristics of neighborhoods. We discuss expansion of our approach to produce synthetic population ensembles for the entire US, particularly plans to establish automated workflows for job orchestration to increase computational efficiency, as well as provide outlook for broadening applications of the ensembles.

Gaboardi, James [ORNL] (ORCID:0000000247766826)↗
DOE OSTI2023

pyAMReX v23.08

The Python binding for AMReX, pyAMReX, bridges the worlds of block-structured codes and data science: it provides zero-copy application GPU data access for AI/ML, in situ analysis, application coupling and enables rapid, massively parallel prototyping.

Huebl, Axel↗
DOE OSTI2020

Pairing Dynamics in Low-energy Nuclear Collisions

Superfluidity is a generic feature of various quantum systems at low temperatures and it is in particular important for the description of dynam- ics of low energy nuclear reactions. The Time-Dependent Density Func- tional Theory (TDDFT) is, to date, the only microscopic method which takes into account in a consistent way far-from-equilibrium dynamics of pairing field and single-particle degrees of freedom. Here, the local version of TDDFT, so-called TDSLDA, is particularly useful for the description of nuclear reactions and is well-suited for leadership class computers of hybrid (CPU+GPU) architecture. The preliminary results obtained for collisions involving both medium-mass and heavy nuclei at the energies around the Coulomb barrier are presented.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2023

Develop a Fast Analysis Solver for Welding Sequence Optimization

During the shipbuilding manufacturing process, materials are exposed to significant stresses, as induced both thermally and mechanically, that alter the intended design and significantly affect the production schedule, labor hours (fitting, welding, rework, etc.), and material structural performance. The type and magnitude of deformation of a given structure depends on many factors such as the material, thickness and quality of components, the process heat input, preheat and inter-pass temperatures, type and size of welds, welding sequence and direction, location, sequence, and degree of fixturing. Numerical simulations using finite element analysis (FEA) have long been used to analyze welding-induced structural distortion. For large assemblies, transient thermal elastic-plastic analysis (TEPA) can take days or weeks to run, and optimization of welding sequence is not feasible. Simplified analysis methods were developed to reduce computational time. However, it is challenging to use these techniques to fully optimize welding sequencing because of their applied simplifications in modeling weld details. A fast analysis solver that could be used by the shipbuilding industry is being developed for optimizing welding sequences by taking full advantage of modern GPU-based HPC hardware and incorporating patented acceleration schemes. The accelerated processing factors are up to 2200 times greater for large, multi-pass welded structures.

Yang, Yu-Ping↗
DOE OSTI2020

Facilitated Transport Membranes with Tunable Amine-CO 2 Chemistry for CO 2 /H 2 Separation

CO 2 -selective, amine-containing facilitated transport membranes are of great interest for syngas purification since high-pressure H 2 can be retained upon CO 2 removal. A variety of amine-containing polymers have shown decent chemical and thermal stability at aggressive conditions, but their CO 2 /H 2 separation properties are largely limited by the hydrostatic compaction and severe carrier saturation associated with the high syngas pressure. Herein, we report a new approach to enhance the CO 2 permeability by manipulating the steric hindrance of the amine carriers. A series of α-aminoacids with different alkyl or hydroxyethyl substituents are deprotonated by 2-(1-piperazinyl)ethylamine, leading to nonvolatile amine carriers with different degrees of steric hindrance. For hosting the low MW amine carriers, a highly crosslinked poly(vinyl acetal) is synthesized as a water-swellable polymer network. In order to avoid membrane compaction, perforated graphene oxide mono-sheets are dispersed as reinforcement fillers. In the presence of moisture, a bulkier alkyl substituent to the amino site (increasing steric hindrance) destabilizes the carbamate adduct and thus drastically increases the chemisorption of CO 2 , while the incorporation of ethylene oxide groups provides additional physisorption of CO 2 . The enhanced CO 2 solubility significantly mitigates the carrier saturation behavior, and an unprecedented CO 2 /H 2 selectivity greater than 100 is demonstrated at 107 °C and 12.5 atm of CO 2 partial pressure. As the CO 2 partial pressure reduces to 0.4 atm, a less hindered amine yields a higher reactive diffusivity of CO 2 , resulting in a CO 2 permeance of 435 GPU with a selectivity greater than 500. These reaction-mediated polymeric membranes are well above the theoretical upper bound, and they are of great interest for designing a highly-selective membrane process for syngas purification.

08 HYDROGEN↗
DOE OSTI2020

Method and apparatus for temperature and voltage management control

A method and apparatus for managing processing power determine a supply voltage to supply to a processing unit, such as a central processing unit (CPU) or graphics processing unit (GPU), based on temperature inversion based voltage, frequency, temperature (VFT) data. The temperature inversion based VFT data includes supply voltages and corresponding operating temperatures that cause the processing unit's transistors to operate in a temperature inversion region. In one example, the temperature inversion based VFT data includes lower supply voltages and corresponding higher temperatures in a temperature inversion region of a processing unit. The temperature inversion based VFT data is based on an operating frequency of the processing unit. The apparatus and method adjust a supply voltage to the processing unit based on the temperature inversion based VFT data.

42 ENGINEERING↗
DOE OSTI2020

MEPHESTO: Modeling Energy-Performance in Heterogeneous SoCs and Their Trade-Offs

Integrated shared memory heterogeneous architectures are pervasive because they satisfy the diverse needs of mobile, autonomous, and edge computing platforms. Although specialized processing units (PUs) that share a unified system memory improve performance and energy efficiency by reducing data movement, they also increase contention for this memory since the PUs interact with each other. Prior work has investigated performance degradation due to memory contention, but few have studied the relationship of power and energy to memory contention. Moreover, a comprehensive solution that models memory contention for kernel placement on contemporary heterogeneous systems on chip (SoCs) in response to energy and performance has been largely unaddressed.This paper presents MEPHESTO, a novel and holistic approach for managing this balance. The authors characterize applications and PUs in terms of two memory contention factors - time factors and power factors - to achieve the desired trade-off between energy and performance for collocated kernel execution on heterogeneous systems. The authors believe that this investigation is the first to combine all of these factors and present a simple knob-based approach that expresses the target trade-off. The approach is evaluated on a diverse integrated shared memory heterogeneous system with a CPU, GPU, and programmable vision accelerator. By using an empirical model for memory contention that provides up to 92% accuracy, the kernel collocation approach can provide a near-optimal ordering and placement based on the user-defined, energy-performance trade-off parameter. Moreover, the dynamic programming-based heuristics provide up to 30% better energy or 20% performance benefits when compared with the greedy approaches commonly employed by previous studies.

Alaul haque monil, Mohammad↗
DOE OSTI2020

Performance Evaluation of Python Based Data Analytics Frameworks in Summit: Early Experiences

The explosion in the volumes of data generated from ever-larger simulation campaigns and experiments or observations necessitates competent tools for data wrangling and analysis). While the Oak Ridge Leadership Computing Facility (OLCF) provides a variety of tools to perform data wrangling and data analysis tasks, Python based tools often lack scalability, or the ability to fully exploit the computational capability of OLCF’s Summit supercomputer. NVIDIA RAPIDS and Dask offer a promising solution to accelerate and distribute data analytics workloads from personal computers to heterogeneous supercomputing systems. We discuss early performance evaluation results of RAPIDS and Dask on Summit to understand their capabilities, scalability, and limitations. Our evaluation includes a subset of RAPIDS libraries, i.e., cuDF, cuML, and cuGraph, and Chainer’s CuPy, and their multi-GPU variants when available.We also draw on the observed trends from the performance evaluation results to discuss best practices for maximizing performance.

Hernandez Arreguin, Benjamin↗
DOE OSTI2020

Big Data Processing for Power Grid Event Detection

In this paper we present the application of big data processing for the development of machine learning (ML) models to detect relevant events in power grid operations. This is based on almost 20TB of phasor measurement unit data corresponding to up to two years of operation of three grid interconnections which provide power to most of the United States. A significant aspect of the work consists in having all data processing performed on a single standard GPU server, from pre-processing to ML model training and testing. We describe the data and computational infrastructure, challenges faced and methods used in data processing, main findings and results. The ML approach employed for best utilization of the big data is also discussed, including sample results.

Paes Leao, Bruno↗
DOE OSTI2021

Porting CMS Heterogeneous Pixel Reconstruction to Kokkos

Programming for a diverse set of compute accelerators in addition to the CPU is a challenge. Maintaining separate source code for each architecture would require lots of effort, and development of new algorithms would be daunting if it had to be repeated many times. Fortunately there are several portability technologies on the market such as Alpaka, Kokkos, and SYCL. These technologies aim to improve the developer productivity by making it possible to use the same source code for many different architectures. In this paper we use heterogeneous pixel reconstruction code from the CMS experiment at the CERNL LHC as a realistic use case of a GPU-targeting HEP reconstruction software, and report experience from prototyping a portable version of it using Kokkos. The development was done in a standalone program that attempts to model many of the complexities of a HEP data processing framework such as CMSSW. We also compare the achieved event processing throughput to the original CUDA code and a CPU version of it.

Childers, Taylor↗
DOE OSTI2020

Generic, Sparse Tensor Core for Neural Networks

Sparse neural network attracts more attention for model compression, fast execution, and power reduction. The state-of-the-art designed sparse tensor core for structured and static sparsity, which did not support well for generic or dynamic sparsity. We design a sparse tensor core to support generic sparsity pruning with a novel hybrid and blocked sparse matrix storage format, HB-ELL, which saves computation and storage while keeping the most significant elements, as well as supporting dynamic sparsity for data flow in neural networks. We achieve better performance with preliminary results than the state-of- the-art on an NVIDIA GPU simulator.

Wu, Xiaolong↗
DOE OSTI2021

Accelerating Transformer-based Deep Learning Models on FPGAs using Column Balanced Block Pruning

Although Transformer-based language representations achieve state-of-the-art accuracy on various natural language processing (NLP) tasks, the large model size has been challenging the resource constrained computing platforms. Weight pruning, as a popular and effective technique in reducing the number of weight parameters and accelerating the Transformer, has been investigated on GPUs. However, the Transformer acceleration using weight pruning on field-programmable gate array (FPGAs) remains unexplored. This paper investigates the column balanced block-wise pruning on Transformer and designs an FPGA acceleration engine to customize the balanced blockwise matrix multiplication. We implement the Transformer model with proper hardware scheduling, and the experiments show that the Transformer inference on FPGA achieves 10.35 ms latency with the batch size of 32, which is 10.96 × speed up comparing to CPU platform and 2.08 × speed up comparing to GPU platform.

Peng, Hongwu↗
DOE OSTI2021

Accelerating Multigrid-based Hierarchical Scientific Data Refactoring on GPUs

Rapid growth in scientific data and a widening gap between computational speed and I/O bandwidth make it increasingly infeasible to store and share all data produced by scientific simulations. Instead, we need methods for reducing data volumes: ideally, methods that can scale data volumes adaptively so as to enable negotiation of performance and fidelity tradeoffs in different situations. Multigrid-based hierarchical data representations hold promise as a solution to this problem, allowing for flexible conversion between different fidelities so that, for example, data can be created at high fidelity and then transferred or stored at lower fidelity via logically simple and mathematically sound operations. However, the effective use of such representations has been hindered until now by the relatively high costs of creating, accessing, reducing, and otherwise operating on such representations. We describe here highly optimized data refactoring kernels for GPU accelerators that enable efficient creation and manipulation of data in multigrid-based hierarchical forms. We demonstrate that our optimized design can achieve up to 250 TB/s aggregated data refactoring throughput—83% of theoretical peak—on 1024 nodes of the Summit supercomputer. We showcase our optimized design by applying it to a large-scale scientific visualization workflow and the MGARD lossy compression software.

Chen, Jieyang↗
DOE OSTI2021

High speed two-dimensional imaging with an analog interface

A quantitative pulse count (event detection) algorithm with linearity to high count rates is accomplished by combining a high-speed, high frame rate camera with simple logic code run on a massively parallel processor such as a GPU. The parallel processor elements examine frames from the camera pixel by pixel to find and tag events or count pulses. The tagged events are combined to form a combined quantitative event image.

Waugh, Justin↗
DOE OSTI2020

The Baryon Mapping Experiment (BMX), a 21cm intensity mapping pathfinder

The Baryon Mapping eXperiment (BMX) is an interferometric array designed as a pathfinder for a future post- reionization 21 cm intensity mapping survey. It consists of four 4-meter parabolic reflectors each having offset pyramidal horn feed, quad-ridge orthomode transducer, temperature-stabilized RF amplification and filtering, and pulsed noise injection diode. An undersampling readout scheme uses 8-bit digitizers running at 1.1 Gsamples/sec to provide access to signals from 1.1 - 1.55 GHz (third Nyquist zone), corresponding to HI emission from sources at redshift 0 < z < 0.3. An FX correlator is implemented in GPU and generates 28 GB/day of time-ordered visibility data. About 7,000 hours of data were collected from Jan. 2019 - May 2020, and we will present results on system performance including sensitivity, beam mapping studies, observations of bright celestial targets, and system electronics upgrades. BMX is a pathfinder for the proposed PUMA intensity mapping survey in the 2030s.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2021

Large scale multi-node simulations of $\mathbb{Z}_2$ gauge theory quantum circuits using Google Cloud Platform

Simulating quantum field theories on a quantum computer is one of the most exciting fundamental physics applications of quantum information science. Dynamical time evolution of quantum fields is a challenge that is beyond the capabilities of classical computing, but it can teach us important lessons about the fundamental fabric of space and time. Whether we may answer scientific questions of interest using near-term quantum computing hardware is an open question that requires a detailed simulation study of quantum noise. Here we present a large scale simulation study powered by a multi-node implementation of qsim using the Google Cloud Platform. We additionally employ newly-developed GPU capabilities in qsim and show how Tensor Processing Units -- Application-specific Integrated Circuits (ASICs) specialized for Machine Learning -- may be used to dramatically speed up the simulation of large quantum circuits. We demonstrate the use of high performance cloud computing for simulating $\mathbb{Z}_2$ quantum field theories on system sizes up to 36 qubits. We find this lattice size is not able to simulate our problem and observable combination with sufficient accuracy, implying more challenging observables of interest for this theory are likely beyond the reach of classical computation using exact circuit simulation.

Gustafson, Erik↗
DOE OSTI2022

m-CUBES An efficient and portable implementation of multi-dimensional integration for gpus

The task of multi-dimensional numerical integration is frequently encountered in physics and other scientific fields, e.g., in modeling the effects of systematic uncertainties in physical systems and in Bayesian parameter estimation. Multi-dimensional integration is often time-prohibitive on CPUs. Efficient implementation on many-core architectures is challenging as the workload across the integration space cannot be predicted a priori. We propose m-Cubes, a novel implementation of the well-known Vegas algorithm for execution on GPUs. Vegas transforms integration variables followed by calculation of a Monte Carlo integral estimate using adaptive partitioning of the resulting space. m-Cubes improves performance on GPUs by maintaining relatively uniform workload across the processors. As a result, our optimized Cuda implementation for Nvidia GPUs outperforms parallelization approaches proposed in past literature. We further demonstrate the efficiency of m-Cubes by evaluating a six-dimensional integral from a cosmology application, achieving significant speedup and greater precision than the CUBA library's CPU implementation of VEGAS. We also evaluate m-Cubes on a standard integrand test suite. m-Cubes outperforms the serial implementations of the Cuba and GSL libraries by orders of magnitude speedup while maintaining comparable accuracy. Our approach yields a speedup of at least 10 when compared against publicly available Monte Carlo based GPU implementations. In summary, m-Cubes can solve integrals that are prohibitively expensive using standard libraries and custom implementations. A modern C++ interface header-only implementation makes m-Cubes portable, allowing its utilization in complicated pipelines with easy to define stateful integrals. Compatibility with non-Nvidia GPUs is achieved with our initial implementation of m-Cubes using the Kokkos framework.

Sakiotis, Ioannis↗