Search NASA⌕ Search

SEARCH · Search NASA

Results for “data parallel”

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 OSTI2017

MPI-FAUN: An MPI-Based Framework for Alternating-Updating Nonnegative Matrix Factorization

Non-negative matrix factorization (NMF) is the problem of determining two non-negative low rank factors W and H, for the given input matrix A, such that A≈WH. NMF is a useful tool for many applications in different domains such as topic modeling in text mining, background separation in video analysis, and community detection in social networks. Despite its popularity in the data mining community, there is a lack of efficient parallel algorithms to solve the problem for big data sets. The main contribution of this work is a new, high-performance parallel computational framework for a broad class of NMF algorithms that iteratively solves alternating non-negative least squares (NLS) subproblems for W and H. It maintains the data and factor matrices in memory (distributed across processors), uses MPI for interprocessor communication, and, in the dense case, provably minimizes communication costs (under mild assumptions). The framework is flexible and able to leverage a variety of NMF and NLS algorithms, including Multiplicative Update, Hierarchical Alternating Least Squares, and Block Principal Pivoting. Our implementation allows us to benchmark and compare different algorithms on massive dense and sparse data matrices of size that spans from few hundreds of millions to billions. We demonstrate the scalability of our algorithm and compare it with baseline implementations, showing significant performance improvements. The code and the datasets used for conducting the experiments are available online.

2D↗
DOE OSTI2024

Efficient Parallelization of Irregular Applications on GPU Architectures

With the enlarging computation capacity of general Graphics Processing Units (GPUs), leveraging GPUs to accelerate parallel applications has become a critical topic in academia and industry. However, a wide range of irregular applications with the computation-/memory-intensive nature cannot easily achieve high GPU utilization. The challenges mainly involve the following aspects: first, data dependence leads to coarse-grained kernel and inefficient parallelism; second, heavy GPU memory usage may cause frequent memory evictions and extra overhead of I/O; third, specific computation patterns produce memory redundancies; last, workload balance and data reusability conjunctly benefit the overall performance, but there may exist a dynamic trade-off between them. Targeting these challenges, this dissertation proposes multiple optimizations to accelerate two real-world applications: many-body correlation functions to simulate nuclear physics in a large-scale scientific system; the other is the eALS-based matrix factorization recommendation system. To accelerate the calculations of many-body correlation functions, this dissertation presents three frameworks in GPU memory management and multi-GPU scheduling. Firstly, an optimized systematic GPU memory management framework, MemHC, utilizes a series of new memory reduction designs in GPU memory allocation, CPU/GPU communications, and GPU memory oversubscription. Secondly, an enhanced multi-GPU scheduling framework, MICCO, particularly by taking both data dimension (e.g., data reuse and data eviction) and computation dimension into account. MICCO designs a heuristic scheduling algorithm and a machine learning-based regression model to generate the optimal settings of a proposed new concept to manage the trade-off. Thirdly, a locality-aware multi-GPU scheduling framework. This scheduler leverages pipeline batch generation with a looking-ahead strategy by building local dependency graphs for memory transfer reduction and better data reuse, achieving up to 79.92% memory cost reduction and 1.67x speedup. To parallelize the eALS-based recommendation system, this dissertation proposes an efficient CPU/GPU heterogeneous recommendation system, HEALS. HEALS employs newly designed architecture-adaptive data formats to achieve load balance and good data locality on CPU and GPU. To mitigate the data dependence, HEALS presents a CPU/GPU collaboration model for both task parallelism and data parallelism with multiple kernel computation optimizations. In summary, this dissertation efficiently accelerates two typical irregular applications on GPUs by building four frameworks, including CPU/GPU collaboration, GPU memory management, and multi-GPU scheduling.

Wang, Qihan↗
DOE OSTI2019

Parsl: Pervasive Parallel Programming in Python

High-level programming languages such as Python are increasingly used to provide intuitive interfaces to libraries written in lower-level languages and for assembling applications from various components. This migration towards orchestration rather than implementation, coupled with the growing need for parallel computing (e.g., due to big data and the end of Moore's law), necessitates rethinking how parallelism is expressed in programs. Here, we present Parsl, a parallel scripting library that augments Python with simple, scalable, and flexible constructs for encoding parallelism. These constructs allow Parsl to construct a dynamic dependency graph of components that it can then execute efficiently on one or many processors. Parsl is designed for scalability, with an extensible set of executors tailored to different use cases, such as low-latency, high-throughput, or extreme-scale execution. We show, via experiments on the Blue Waters supercomputer, that Parsl executors can allow Python scripts to execute components with as little as 5 ms of overhead, scale to more than 250000 workers across more than 8000 nodes, and process upward of 1200 tasks per second. Other Parsl features simplify the construction and execution of composite programs by supporting elastic provisioning and scaling of infrastructure, fault-tolerant execution, and integrated wide-area data management. We show that these capabilities satisfy the needs of many-task, interactive, online, and machine learning applications in fields such as biology, cosmology, and materials science.

parallel programming↗
DOE OSTI2025

2D reactive transport model of shale chemical weathering and biogeochemical fluxes along a mountainous hillslope, East River Watershed, Colorado: Input files and simulation results

This data package contains input files and simulation results for a two-dimensional (2D) reactive transport model used to quantitatively analyze the coupled hydrological and biogeochemical processes governing shale weathering and associated biogeochemical fluxes under realistic environmental conditions in the high-elevation East River Watershed. These data support the conclusions presented in Stolze et al. (Water Resources Research, under review), "Model-based interpretation of solute exports and carbon partitioning during shale weathering in a mountainous hillslope". The model simulates atmospheric-subsurface gas exchange, subsurface water flow, and shale weathering processes under dynamic, year-scale conditions along a shale-underlain hillslope located in the East River watershed. The simulations were performed using the PFLOTRAN flow and reactive transport code and executed on the Perlmutter supercomputer to leverage its large-scale parallel computing capabilities. The data package contains two zipped folders, "model_input_files" and "simulation_results", and one readme.txt file. "model_input_files" contains the necessary input files to run the calibrated base-base model presented in Stolze et al. (Water Resources Research, under review). "simulation_results" contains a single hdf5 file ("Output_2D_hillslope_model.h5") which includes the results of simulation performed using the base-case model. This file can be opened with HDFView 3.1.4, Python, or MATLAB. "readme.txt" contains relevant information about the base-case model and provides guidelines on how to run the associated input files provided in the folder "model_input_files". Furthermore, readme.txt provides information regarding the model results provided in "Output_2D_hillslope_model.h5" such as matrix dimensionality and output units. Field datasets used to evaluate model performance were collected at three monitoring wells located along a hillslope transect (PLM1, PLM2, and PLM3). Dissolved ion concentration data were collected from November 2016 to October 2021 for Ca, Mg, DIC, Na, K, SO4 (Dong et al., 2025 - dic_npoc_data_2014_2024.zip - DOI:10.15485/1660459; Williams et al., 2025 - anion_data_2014_2024.zip - DOI:10.15485/1668054; Dong et al., 2025 - cation_data_2014_2024.zip - DOI:10.15485/1668055). Note that we used the files named er_PLM1_xx_yy, er_PLM2_xx_yy, and er_PLM3_xx_yy where xx stands for the name of the aqueous species and yy stands for the depth where the measurements were performed. Soil water content ([0 - 1] m) and water table depth were collected from November 2016 to October 2021 (Wan et al., 2024 - Dynamic_water_table__depthsFig2b.csv and Soil_water_content_Fig4e.csv - DOI:10.15485/2322567). Gaseous CO2 concentration were collected from October 2020 to December 2021(Wan et al., 2024 - Soil_CO2_concentrations_Fig4h.csv - DOI:10.15485/2322567) Gaseous CO2 flux from the subsurface to the atmosphere were collected in the vicinity of PLM2 from October 2019 to May 2022 (Wu et al., 2025). Soil microbial biomass concentration was measured from August 2016 to June 2017 (Sorensen et al., 2019 - 2017_East_River_Pumphouse_Microbial_Biomass__1_.csv - DOI:10.15485/1577267) All field data are published as CSV files compatible with Microsoft Excel, MATLAB, and Python, or as text files. The coordinates of the monitoring wells and the CO2(g) flux sensor in the coordinate system WGS84 are: -PLM1: [38.9197710 ; -106.9492750] -PLM2: [38.9201580 ; -106.9487170] -PLM3: [38.9207843 ; -106.9483668] -PLM4: 38.9210060 ; -106.9479528] -CO2(g) flux sensor: [38.9199180 ; -106.9489906] ------------------------------------------------------------------------------------------- This work was supported by the Watershed Function Science Focus Area at Lawrence Berkeley National Laboratory funded by the US Department of Energy, Office of Science, Biological and Environmental Research under Contract No. DE-AC02-05CH11231. This research used resources of the National Energy Research Scientific Computing Center (NERSC), a Department of Energy User Facility using NERSC award BER-ERCAP 23980, BER-ERCAP 28550, and BER-ERCAP 33789.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2017

Building Halo Merger Trees from the Q Continuum Simulation

Cosmological N-body simulations rank among the most computationally intensive efforts today. A key challenge is the analysis of structure, substructure, and the merger history for many billions of compact particle clusters, called halos. Effectively representing the merging history of halos is essential for many galaxy formation models used to generate synthetic sky catalogs, an important application of modern cosmological simulations. Generating realistic mock catalogs requires computing the halo formation history from simulations with large volumes and billions of halos over many time steps, taking hundreds of terabytes of analysis data. We present fast parallel algorithms for producing halo merger trees and tracking halo substructure from a single-level, density-based clustering algorithm. Merger trees are created from analyzing the halo-particle membership function in adjacent snapshots, and substructure is identified by tracking the "cores" of merging halos - sets of particles near the halo center. Core tracking is performed after creating merger trees and uses the relationships found during tree construction to associate substructures with hosts. The algorithms are implemented with MPI and evaluated on a Cray XK7 supercomputer using up to 16,384 processes on data from HACC, a modern cosmological simulation framework. We present results for creating merger trees from 101 analysis snapshots taken from the Q Continuum, a large volume, high mass resolution, cosmological simulation evolving half a trillion particles.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2017

Trace: a high-throughput tomographic reconstruction engine for large-scale datasets

Abstract Background Modern synchrotron light sources and detectors produce data at such scale and complexity that large-scale computation is required to unleash their full power. One of the widely used imaging techniques that generates data at tens of gigabytes per second is computed tomography (CT). Although CT experiments result in rapid data generation, the analysis and reconstruction of the collected data may require hours or even days of computation time with a medium-sized workstation, which hinders the scientific progress that relies on the results of analysis. Methods We present Trace, a data-intensive computing engine that we have developed to enable high-performance implementation of iterative tomographic reconstruction algorithms for parallel computers. Trace provides fine-grained reconstruction of tomography datasets using both (thread-level) shared memory and (process-level) distributed memory parallelization. Trace utilizes a special data structure called replicated reconstruction object to maximize application performance. We also present the optimizations that we apply to the replicated reconstruction objects and evaluate them using tomography datasets collected at the Advanced Photon Source. Results Our experimental evaluations show that our optimizations and parallelization techniques can provide 158× speedup using 32 compute nodes (384 cores) over a single-core configuration and decrease the end-to-end processing time of a large sinogram (with 4501 × 1 × 22,400 dimensions) from 12.5 h to <5 min per iteration. Conclusion The proposed tomographic reconstruction engine can efficiently process large-scale tomographic data using many compute nodes and minimize reconstruction times.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2022

Ponderomotive force driven density modifications parallel to B on the LAPD

The ponderomotive force has previously been identified as a possible driver of observed density modifications close to radio frequency actuators during operation. This nonlinear force redistributes density in regions of gradients in the magnitude of an oscillating electric field and describes the influence of the fast time scale RF wave dynamics on slow time scale plasma transport. Depletion of the saturation current (a proxy for the density) measured at the Large Plasma Device (LAPD) was 30–35% during ion cyclotron range of frequencies operation. A coupled 1D plasma transport and cold plasma frequency domain wave solver was developed to self-consistently describe ponderomotive effects and was used to compare with results obtained from the LAPD experiment. The scaled current density driver for the wave model yielded an RF B field in close agreement with two components of the experimental data. However, the 1D parallel model did not accurately reproduce the amplitude or spatial distribution observed in experimental measurements of By. Within the limitations of the 1D model, initial simulation results showed that the ponderomotive force depleted up to 8% for high power (1 MW) and around 1% for the experimental power of 120 kW. This could suggest that the ponderomotive force is not the main driver of density modification for the LAPD experiments presented in this paper. Higher fidelity tools of at least 2D will be required to give a more realistic description of the RF E fields and the effect of the ponderomotive force on the LAPD.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2016

PolyCheck: Dynamic Verification of Iteration Space Transformations on Affine Programs

High-level compiler transformations, especially loop transformations, are widely recognized as critical optimizations to restructure programs to improve data locality and expose parallelism. Guaranteeing the correctness of program transformations is essential, and to date three main approaches have been developed: proof of equivalence of affine programs, matching the execution traces of programs, and checking bit-by-bit equivalence of the outputs of the programs. Each technique suffers from limitations in either the kind of transformations supported, space complexity, or the sensitivity to the testing dataset. In this paper, we take a novel approach addressing all three limitations to provide an automatic bug checker to verify any iteration reordering transformations on affine programs, including non-affine transformations, with space consumption proportional to the original program data, and robust to arbitrary datasets of a given size. We achieve this by exploiting the structure of affine program control- and data-flow to generate at compile-time lightweight checker code to be executed within the transformed program. Experimental results assess the correctness and effectiveness of our method, and its increased coverage over previous approaches.

affine programs↗
DOE OSTI2017

A multi-platform evaluation of the randomized CX low-rank matrix factorization in Spark

We investigate the performance and scalability of the randomized CX low-rank matrix factorization and demonstrate its applicability through the analysis of a 1TB mass spectrometry imaging (MSI) dataset, using Apache Spark on an Amazon EC2 cluster, a Cray XC40 system, and an experimental Cray cluster. We implemented this factorization both as a parallelized C implementation with hand-tuned optimizations and in Scala using the Apache Spark high-level cluster computing framework. We obtained consistent performance across the three platforms: using Spark we were able to process the 1TB size dataset in under 30 minutes with 960 cores on all systems, with the fastest times obtained on the experimental Cray cluster. In comparison, the C implementation was 21X faster on the Amazon EC2 system, due to careful cache optimizations, bandwidth-friendly access of matrices and vector computation using SIMD units. We report these results and their implications on the hardware and software issues arising in supporting data-centric workloads in parallel and distributed environments.

Gittens, Alex↗
DOE OSTI2022

Toward designing effective exascale scientific computing workflows: experiences and best practices

Many fields within scientific computing have embraced advances in big-data analysis and machine learning, which often requires the deployment of large, distributed and complicated workflows that may combine training neural networks, performing simulations, running inference, and performing database queries and data analysis in asynchronous, parallel and pipelined execution frameworks. Such a shift has brought into focus the need for scalable, efficient workflow management solutions with reproducibility, error and provenance handling, traceability, and checkpoint-restart capabilities, among other needs. Here, we discuss challenges and best-practices for deploying exascale-generation computational science workflows on resources at the Oak Ridge Leadership Computing Facility (OLCF). We present our experiences with large-scale deployment of distributed workflows on the Summit supercomputer, including for bioinformatics and computational biophysics, materials science, and deep learning model optimization. We also present problems and solutions created by working within a Python-centric software base on traditional HPC systems, and discuss steps that will be required before the convergence of HPC, AI, and data science can be fully realized. Our results point to a wealth of exciting new possibilities for harnessing this convergence to tackle new scientific challenges.

Coletti, Mark↗
DOE OSTI2021

Interactive Web Application for Traffic Simulation Data Management and Visualization

As traffic simulation software becomes more effective for realistically simulating and analyzing traffic dynamics and vehicle interactions on the mesoscopic and microscopic level, the management, dissemination, and collaborative visualization of traffic simulation results produced by individual transportation planners presents a significant challenge. Existing online content management systems have a very limited capability in allowing users to query specific traffic simulation scenarios and geospatially visualize simulation results through shareable and interactive web interfaces. This paper presents a web-based application for promoting the archiving, sharing, and visualization of large-scale traffic simulation outputs. The application is developed to enhance cyber-physical controls, communications, and public education for collaborative transportation planning. Unique features of the web application include: (a) allowing users to upload their new traffic simulation scenarios (parameters and outputs), as well as search existing scenarios using easily accessible interfaces; (b) optimizing simulation output files with heterogeneous data formats and projected coordinate systems for web-based storage and management using a scalable and searchable data/metadata standard; (c) standardizing user-uploaded simulation outputs using web interfaces and data processing libraries with parallel computing capacity; and (d) providing shareable web visual interfaces for visualizing the traffic flow and signal information stored in simulation outputs (e.g., regional traffic patterns and individual vehicle interactions) and visually comparing multiple simulation outputs both spatially and temporally. Furthermore, the paper presents the conceptual design and implementation of this application, and demonstrates the application’s performance for sharing, comparing, and visualizing simulation outputs from VISSIM and SUMO, two commonly used traffic simulation software programs.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗
DOE OSTI2022

Measurement of the 240 Pu(n,f) Prompt Fission Neutron Spectra with Chi-Nu

The DOE Nuclear Criticality Safety Program has funded a multi-year effort to measure the 240 Pu(n,f) Prompt Fission Neutron Spectrum at Chi-Nu. This project is a joint LANL-LLNL effort, involving the preparation of 12 240 Pu foils and construction of a Parallel-Plate Avalanche Counter at LLNL, with subsequent measurements of the 240 Pu(n,f) PFNS at the Chi-Nu beamline at the Los Alamos Neutron Science Center’s Weapons Neutron Research facility (LANSCE/WNR). We expect the PPAC to be ready in time to perform the measurements this run cycle.

240Pu Foils↗
DOE OSTI2019

A Scalable Hybrid Scheme for Ray-Casting of Unstructured Volume Data

We present an algorithm for parallel volume rendering that is a hybrid between classical object order and image order techniques. The algorithm operates on unstructured grids (and structured ones), and thus can deal with block boundaries interleaving in complex ways. It also deals effectively with cases that are prone to load imbalance, i.e., cases where cell sizes differ dramatically, either because of the nature of the input data, or because of the effects of the camera transformation. The algorithm divides work over resources such that each phase of its processing is bounded in the amount of computation it can perform. We demonstrate its efficacy through a series of studies, varying over camera position, data set size, transfer function, image size, and processor count. At its biggest, our experiments scaled up to 8,192 processors and operated on data sets with more than one billion cells. In total, we find that our hybrid algorithm performs well in all cases. This is because our algorithm naturally adapts its computation based on workload, and can operate like either an object order technique or an image order technique in scenarios where those techniques are efficient.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Productive Programming of Distributed Systems with the SHAD C++ Library

High-performance computing (HPC) is often perceived as a matter of making large-scale systems (e.g., clusters) run as fast as possible, regardless the required programming effort. However, the idea of "bringing HPC to the masses" has recently emerged. Inspired by this vision, we have designed SHAD, the Scalable High-performance Algorithms and Data-structures library. SHAD is open source software, written in C++, for C++ developers. Unlike other HPC libraries for distributed systems, which rely on SPMD models, SHAD adopts a shared-memory programming abstraction, to make C++ programmers feel at home. Underneath, SHAD manages tasking and data-movements, moving the computation where data resides and taking advantage of asynchrony to tolerate network latency. At the bottom of his stack, SHAD can interface with multiple runtime systems: this not only improves developer’s productivity, by hiding the complexity of such software and of the underlying hardware, but also greatly enhance code portability. Thanks to its abstraction layers, SHAD can indeed target different systems, ranging from laptops to HPC clusters, without any need for modifying the user-level code. We have prototyped and open-sourced the implementation of (a subset of) the C++ standard library (STL) targeting multi-node HPC clusters. Our work allows plain STL-based C++ code to scale on HPC systems, with no need for rewriting the code to exploit the complex hardware. SHAD is available under Apache v2 License at https://github.com/pnnl/SHAD. In this paper we overview the design of the SHAD library, depicting its main components: runtime systems abstractions for tasking; parallel and distributed data-structures; STL-compliant interfaces and algorithms.

Castellana, Vito G.↗
DOE OSTI2016

Asynchronous in situ connected-components analysis for complex fluid flows

Fluid flow in porous media is at the heart of problems such as groundwater contamination and carbon sequestration, and presents an important challenge for scientific computing. For this class of problem, large three-dimensional simulations are performed to advance scientific inquiry. On massively parallel computing systems, the volume of data generated by such approaches can become a productivity bottleneck if the raw data generated from the simulation is analyzed in a post-processing step. We present a physics-based framework for in situ data reduction that is theoretically grounded in multiscale averaging theory. We show how task parallelism can be exploited to concurrently perform a variety of analysis tasks with data-dependent costs, including the generation of iso-surfaces, morphological analyses, and connected components analysis. A task management framework is constructed to launch asynchronous analysis threads, manage dependencies between different tasks, promote data locality and hide the impact of data transfers. The framework is applied to analyze GPU-based simulations of two-fluid flow in porous media, generating a set of averaged measures that represents the overall system behavior. We demonstrate how the approach can be applied to perform physically-consistent averaging over fluid subregions using connected components analysis. Simulations performed on Oak Ridge National Lab s Titan supercomputer are profiled to demonstrate the performance of the associated multi-threaded in situ analysis for typical production simulation of two-fluid flow.

Mcclure, James↗
DOE OSTI2018

A comment on the position dependent diffusion coefficient representation of structural heterogeneity

Experimental studies of the variation of the mean square displacement (MSD) of a particle in a confined colloid suspension that exhibits density variations on the scale length of the particle diameter are not in agreement with the prediction that the spatial variation in MSD should mimic the spatial variation in density. The predicted behavior is derived from the expectation that the MSD of a particle depends on the system density and the assumption that the force acting on a particle is a point function of position. The experimental data are obtained from studies of the MSDs of particles in narrow ribbon channels and between narrowly spaced parallel plates and from new data, reported herein, of the radial and azimuthal MSDs of a colloid particle in a dense colloid suspension confined to a small circular cavity. In each of these geometries, a dense colloid suspension exhibits pronounced density oscillations with spacing of a particle diameter. We remove the discrepancy between prediction and experiment using the Fisher-Methfessel interpretation of how local equilibrium in an inhomogeneous system is maintained to argue that the force acting on a particle is delocalized over a volume with radius equal to a particle diameter. Our interpretation has relevance to the relationship between the scale of inhomogeneity and the utility of translation of the particle MSD into a position dependent diffusion coefficient and to the use of a spatially dependent diffusion coefficient to describe mass transport in a heterogeneous system.

74 ATOMIC AND MOLECULAR PHYSICS↗
DOE OSTI2021

FasTensor (FT) v0.0.1

The FasTensor implements the native array data programming mode namely SLOPE for modern data analysis tasks. It is designed to work in parallel on supercomputer and can handle terabyte data analysis tasks. It is thousand times faster than other data analysis systems such as Apache Spark and 38% faster than TensorFlow in dealing with large scale scientific data analysis. It has wide applications in image data analysis and mesh data analysis tasks in physical simulations and other fields.

Wu, Kesheng↗
DOE OSTI2023

Developing Ultrahigh-Resolution E3SM Land Model for GPU Systems

Designing and refactoring complex scientific code, such as the E3SM land model (ELM), for new computing architectures is challenging. This paper presents design strategies and technical approaches to develop a data-oriented, GPU-ready ELM model using compiler directives (OpenACC/OpenMP). We first analyze the datatypes and processes in the original ELM code. Then we present design considerations for ultrahigh-resolution ELM (uELM) development for massive GPU systems. These techniques include the global data-oriented simulation workflow, domain partition, code porting and data copy, memory reduction, parallel loop restructure and flattening, and race condition detection. We implemented the first version of uELM using OpenACC targeting the NVidia GPUs in the Summit supercomputer at Oak Ridge National Laboratory. During the implementation, we developed a software tool (named SPEL) to facilitate code generation, verification, and performance tuning using these techniques. The first uELM implementation for Nvidia GPUs on Summit delivered promising results: 1) over 98% of the ELM code was automatically generated and tuned by scripts. Most ELM modules had better computational performances than the original ELM code for CPUs. The GPU-ready uELM is more scalable than the CPU code on fully-loaded Summit nodes. Example profiling results from several modules are also presented to illustrate the performance improvements and race condition detection. The lessons learned and toolkit developed in the study are also suitable for further uELM deployment using OpenMP on the first US exascale computer, Frontier, equipped with AMD CPUs and GPUs.

Schwartz, Peter↗