Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Data parallelism”

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 73 records · Page 4
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

EdgeAI: Machine learning via direct attached accelerator for streaming data processing at high shot rate x-ray free-electron lasers

We present a case for low batch-size inference with the potential for adaptive training of a lean encoder model. We do so in the context of a paradigmatic example of machine learning as applied in data acquisition at high data velocity scientific user facilities such as the Linac Coherent Light Source-II x-ray Free-Electron Laser. We discuss how a low-latency inference model operating at the data acquisition edge can capitalize on the naturally stochastic nature of such sources. We simulate the method of attosecond angular streaking to produce representative results whereby simulated input data reproduce high-resolution ground truth probability distributions. By minimizing the mean-squared error between the decoded output of the latent representation and the ground truth distributions, we ensure that the encoding layers and resulting latent representation maintains full fidelity for any downstream task, be it classification or regression. We present throughput results for data-parallel inference of various batch sizes, some with throughput exceeding 100 k images per second. We also show in situ training below 10 s per epoch for the full encoder–decoder model as would be relevant for streaming and adaptive real-time data production at our nation’s scientific light sources.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

FLEET: Flexible Efficient Ensemble Training for Heterogeneous Deep Neural Networks

Parallel training of an ensemble of Deep Neural Networks (DNN) on a cluster of nodes is an effective approach to shorten the process of neural network architecture search and hyper-parameter tuning for a given learning task. Prior efforts have shown that data sharing, where the common preprocessing operation is shared across the DNN training pipelines, saves computational resources and improves pipeline efficiency. Data sharing strategy, however, performs poorly for a heterogeneous set of DNNs where each DNN has varying computational needs and thus different training rate and convergence speed. This paper proposes FLEET, a flexible ensemble DNN training framework for efficiently training a heterogeneous set of DNNs. We build FLEET via several technical innovations. We theoretically prove that an optimal resource allocation is NP-hard and propose a greedy algorithm to efficiently allocate resources for training each DNN with data sharing. We integrate data-parallel DNN training into ensemble training to mitigate the differences in training rates and introduce checkpointing into this context to address the issue of different convergence speeds. Experiments show that FLEET significantly improves the training efficiency of DNN ensembles without compromising the quality of the result.

Guan, Hui↗
DOE OSTI2025

A Versatile Simulated Data Transport Layer for in Situ Workflows Performance Evaluation

In situ processing does not only allow scientific applications to face the explosion in data volume and velocity but also to address the time constraints of many simulation-analysis workflows by providing scientists with early insights about their applications at runtime. Multiple frameworks implement the concept of a data transport layer (DTL) to enable such in situ workflows. These tools are very versatile, directly or indirectly access the data generated on the same node, another node of the same compute cluster, or a completely distinct node, and allow data publishers and subscribers to run on the same computing resources or not. This versatility puts on researchers the onus of taking key decisions related to resource allocation and how to transport data to ensure the most efficient execution of their in situ workflows. However, domain scientists and workflow practitioners lack the appropriate tools to assess the respective performance of particular design and deployment options. In this paper we introduce a versatile simulated DTL designed to provide researchers with insights on the respective performance of different execution scenarios of in situ workflows. This open-source, standalone library builds on the SimGrid toolkit and can be linked to any SimGrid-based simulator. It facilitates the evaluation of the performance behavior, at scale, of different data transport configurations and the study of the effects of resource allocation strategies. We demonstrate the scalability, versatility, and accuracy of this simulated DTL by reproducing the execution of two synthetic benchmarks and of a real-world in situ workflow composed of an MPI application and a parallel data analysis. Results of simulations run on a single core show that the proposed library can simulate the interactions of tens of thousands of simulated processes deployed on two interconnected commodity clusters in a few seconds, and the execution by a thousand simulated processes of an in situ workflow in less than three minutes.

Suter, Fred [ORNL] (ORCID:0000000319021955)↗
DOE OSTI2021

Virtual Log-Structured Storage for High-Performance Streaming

Over the past decade, given the higher number of data sources (e.g., Cloud applications, Internet of things) and critical business demands, Big Data transitioned from batch-oriented to real-time analytics. Stream storage systems, such as Apache Kafka, are well known for their increasing role in real-time Big Data analytics. For scalable stream data ingestion and processing, they logically split a data stream topic into multiple partitions. Stream storage systems keep multiple data stream copies to protect against data loss while implementing a stream partition as a replicated log. This architectural choice enables simplified development while trading cluster size with performance and the number of streams optimally managed. This paper introduces a shared virtual log-structured storage approach for improving the cluster throughput when multiple producers and consumers write and consume in parallel data streams. Stream partitions are associated with shared replicated virtual logs transparently to the user, effectively separating the implementation of stream partitioning (and data ordering) from data replication (and durability). We implement the virtual log technique in the KerA stream storage system. When comparing with Apache Kafka, KerA improves the cluster ingestion throughput by up to 4x when multiple producers write over hundreds of data streams.

consistent stream ordering↗
DOE OSTI2022

Enabling real-time adaptation of machine learning models at x-ray Free Electron Laser facilities with high-speed training optimized computational hardware

The emergence of novel computational hardware is enabling a new paradigm for rapid machine learning model training. For the Department of Energy’s major research facilities, this developing technology will enable a highly adaptive approach to experimental sciences. In this manuscript we present the per-epoch and end-to-end training times for an example of a streaming diagnostic that is planned for the upcoming high-repetition rate x-ray Free Electron Laser, the Linac Coherent Light Source-II. We explore the parameter space of batch size and data parallel training across multiple Graphics Processing Units and Reconfigurable Dataflow Units. We show the landscape of training times with a goal of full model retraining in under 15 min. Although a full from scratch retraining of a model may not be required in all cases, we nevertheless present an example of the application of emerging computational hardware for adapting machine learning models to changing environments in real-time, during streaming data acquisition, at the rates expected for the data fire hoses of accelerator-based user facilities.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Synchronous High-frequency Distributed Readout For Edge Processing At The Fermilab Main Injector And Recycler

The Main Injector (MI) was commissioned using data acquisition systems developed for the Fermilab Main Ring in the 1980s. New VME-based instrumentation was commissioned in 2006 for beam loss monitors (BLM)[2], which provided a more systematic study of the machine and improved displays of routine operation. However, current projects are demanding more data and at a faster rate from this aging hardware. One such project, Real-time Edge AI for Distributed Systems (READS), requires the high-frequency, low-latency collection of synchronized BLM readings from around the approximately two-mile accelerator complex. Significant work has been done to develop new hardware to monitor the VME backplane and broadcast BLM measurements over Ethernet, while not disrupting the existing operations critical functions of the BLM system. This paper will detail the design, implementation, and testing of this parallel data pathway.

43 PARTICLE ACCELERATORS↗
DOE OSTI2025

Knowledge Oriented Graph Unified Transformer (KOGUT) v0.1

KOGUT — Knowledge Oriented Graph Unified Transformer KOGUT implements the Relational Graph Transformer (RelGT) architecture for knowledge graph link prediction in biological domains, with a primary focus on microbial growth media prediction. While the original RelGT (arXiv:2505.10960) targets relational tables, time series, and multi-table databases, KOGUT adapts this architecture for heterogeneous biological knowledge graphs, providing first-in-class AI predictive models for microbial cultivation. Key Adaptations Beyond Original RelGT: - Knowledge Graph Focus: Applied to biological KGs with semantic node types (taxa, chemicals, media, phenotypes, environments) versus generic relational database tables, trained on the KG-Microbe knowledge graph (1.3M entities, 2.9M edges, 24 relation types). - Multimodal Node Encoding: Integrates node labels, categories, descriptions, and synonyms from KG metadata through learned embedding layers—adapting relational column features to graph node attributes with textual semantics. - Extended K-Hop Subgraph Strategy: Optimized neighborhood sampling (3-hop default, configurable up to 200 nodes) tuned for sparse biological networks, building on the original local-global attention framework with biological relation preservation. - Biolink Predicate Preservation: Type-specific transformations for 24 biological edge semantics (occurs_in, consumes, produces, has_phenotype, subclass_of) beyond standard relational foreign keys, enabling multi-relation link prediction. - Inductive Learning Support: Enables zero-shot predictions for novel taxa through feature-based embeddings (temperature, oxygen requirements, gram stain, cell shape), extending the original transductive relational benchmark scope to uncultured microorganisms. CheapSOTA Performance Optimizations (This Distribution): - VQ-EMA Centroid Attention: Vector quantization with exponential moving average for improved global context modeling (+5-10% MRR improvement). - HDF5 Precomputed Data Loading: One-time preprocessing of k-hop subgraphs to eliminate redundant graph traversals (2-5× training speedup). - Distributed Data Parallel Training: Multi-GPU support for scaling to larger knowledge graphs (tested on 4× NVIDIA A100 GPUs at NERSC Perlmutter). - Mixed Precision Training: Automatic mixed precision (AMP) for memory efficiency and faster training. Advantages Over Standard Knowledge Graph Embedding Models: Combines RelGT's proven multi-element tokenization (features, type, hop, structure) with graph-native biological representations, enabling interpretable link prediction across heterogeneous entities that standard embedding models (TransE, RotatE, ComplEx) and table-based transformers cannot directly model. Achieves near-perfect performance on microbial growth media prediction (MRR: 0.9966, Precision@1: 0.9932, Hit@10: 1.0000) while maintaining explainability through attention-based reasoning over biological pathways. Training Data: - KG-Microbe merged knowledge graph: 1,379,337 nodes, 2,960,472 edges - 24 biological relation types including taxonomic hierarchies, metabolic interactions, phenotype associations, and environmental relationships - Primary prediction task: Growth media suitability for microbial taxa (biolink:occurs_in, 50K edges) - Multi-relation capability: Predicts links for any of the 24 relation types, including chemical consumption/production, phenotype associations, and taxonomic classification Citation: Original RelGT Architecture: Dwivedi et al., "Relational Graph Transformer", arXiv:2505.10960, 2025 KOGUT Implementation: Knowledge Oriented Graph Unified Transformer for Microbial Growth Media Prediction Developed at Lawrence Berkeley National Laboratory (LBNL) Trained on NERSC Perlmutter supercomputer

Joachimiak, Marcin [Lawrence Berkeley National Lab↗
DOE OSTI2020

DeepFreeze: Towards Scalable Asynchronous Checkpointing of Deep Learning Models

In the age of big data, deep learning has emerged as a powerful tool to extract insight and exploit its value, both in industry and scientific applications. One common pattern emerging in such applications is frequent checkpointing of the state of the learning model during training, needed in a variety of scenarios: analysis of intermediate states to explain features and correlations with training data, exploration strategies involving alternative models that share a common ancestor, knowledge transfer, resilience, etc. However, with increasing size of the learning models and popularity of distributed data-parallel training approaches, simple checkpointing techniques used so far face several limitations: low serialization performance, blocking I/O, stragglers due to the fact that only a single process is involved in checkpointing. This paper proposes a checkpointing technique specifically designed to address the aforementioned limitations, introducing efficient asynchronous techniques to hide the overhead of serialization and I/O, and distribute the load over all participating processes. Experiments with two deep learning applications (CANDLE and ResNet) on a pre-Exascale HPC platform (Theta) shows significant improvement over state-of-art, both in terms of checkpointing duration and runtime overhead.

checkpointing↗
DOE OSTI2024

A High-Performance Discrete-Element Framework for Simulating Flow and Jamming of Moisture Bearing Biomass Feedstocks

We developed and verified a high-performance open-source discrete element method (DEM) solver with simultaneously-supported feedstock-specific interaction models, including bonded-sphere, liquid bridge, cohesion, and non-linear contact models. Our solver uses parallel data structures on hybrid central and graphics processing unit (CPU/GPU) architectures, with favorable strong scaling performance observed for large problem sizes comprised of (100 M particles), and 4X single-node GPU speedup. The particles for corn stover feedstock were conceptualized and calibrated based on experimental measurements and results. Sensitivity analyses demonstrate that the mass flow rate from a wedge hopper is governed primarily by moisture content, friction coefficient, and cohesion energy density. The model is used to reproduce experimentally observed hopper jamming results, highlighting that the experimental no-flow trends can only be achieved by using non-spherical particles, liquid bridge and cohesion models, highlighting the importance of using concurrent feedstock specialized models for the effective representation of biomass material handling problems.

bioenergy↗
DOE OSTI2021

Grid-based minimization at scale: Feldman-Cousins corrections for light sterile neutrino search

High Energy Physics (HEP) experiments generally employ sophisticated statistical methods to present results in searches of new physics. In the problem of searching for sterile neutrinos, likelihood ratio tests are applied to short-baseline neutrino oscillation experiments to construct confidence intervals for the parameters of interest. The test statistics of the form Δχ2 is often used to form the confidence intervals, however, this approach can lead to statistical inaccuracies due to the small signal rate in the region-of-interest. In this paper, we present a computational model for the computationally expensive Feldman-Cousins corrections to construct a statistically accurate confidence interval for neutrino oscillation analysis. The program performs a grid-based minimization over oscillation parameters and is written in C++. Our algorithms make use of vectorization through Eigen3, yielding a single-core speed-up of 350 compared to the original implementation, and achieve MPI data parallelism by employing DIY. We demonstrate the strong scaling of the application at High-Performance Computing (HPC) sites. We utilize HDF5 along with HighFive to write the results of the calculation to file.

Wospakrik, Marianette↗
DOE OSTI2023

All-electron APW+ lo calculation of magnetic molecules with the SIRIUS domain-specific package

We report APW+lo (augmented plane wave plus local orbital) density functional theory (DFT) calculations of large molecular systems using the domain specific SIRIUS multi-functional DFT package. The APW and FLAPW (full potential linearized APW) task and data parallelism options and the advanced eigen-system solver provided by SIRIUS can be exploited for performance gains in ground state Kohn–Sham calculations on large systems. This approach is distinct from our prior use of SIRIUS as a library backend to another APW+lo or FLAPW code. We benchmark the code and demonstrate performance on several magnetic molecule and metal organic framework systems. We show that the SIRIUS package in itself is capable of handling systems as large as a several hundred atoms in the unit cell without having to make technical choices that result in the loss of accuracy with respect to that needed for the study of magnetic systems.

Chemistry↗
DOE OSTI2021

Characterization of a Pixelated Cadmium Telluride Detector System Using a Polychromatic X-Ray Source and Gold Nanoparticle-Loaded Phantoms for Benchtop X-Ray Fluorescence Imaging

In this paper, the imaging dose and scan time have been considered as the two major constraints for routine benchtop x-ray fluorescence computed tomography (XFCT) imaging. One way to address this issue is to acquire x-ray fluorescence (XRF) signals in parallel through a 2D array of single-crystal detectors or a pixelated detector along with the cone-beam x-ray source. To identify a detector system suitable for this purpose, a commercially available, fully spectroscopic cadmium telluride (CdTe) pixelated detector, HEXITEC (High-Energy X-ray Imaging Technology), was tested under the experimental conditions optimized for benchtop XFCT imaging of gold nanoparticles (GNPs). Specifically, two different parallel-hole stainless steel collimators were fabricated and coupled with the detector for seamless integration into our existing benchtop cone-beam XFCT system. After the detector deployment, this benchtop XFCT system was used to detect XRF photons from GNP-loaded phantoms. A pixel-merging algorithm was introduced to enhance the sensitivity of XRF photon detection thereby minimizing the scan time. The effect of pixel-level charge sharing correction algorithms was investigated within the context of benchtop XFCT imaging. The detector energy resolution, in terms of the full width at half maximum (FWHM) values at different gold K-shell XRF energies, was also determined. Of the two charge sharing correction algorithms examined, the charge sharing addition gave better sensitivity than the charge sharing discrimination (csd). On the other hand, under the current experimental conditions, the energy resolution of the HEXITEC detector was the best with the csd and estimated to be 1.56 keV FWHM at 66-69 keV photon energy. Overall, despite some degradation of the detector energy resolution (compared with typical single crystal CdTe detectors), the HEXITEC detector enabled parallel data acquisition under the experimental conditions typical of benchtop XFCT imaging and operated well within our benchtop XFCT setup.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2022

VWC-BERT: Scaling Vulnerability–Weakness–Exploit Mapping on Modern AI Accelerators

Defending cybersystems needs accurate mapping of software and hardware vulnerabilities to generalized descriptions of weaknesses, and weaknesses to exploits. These mappings enable cyber defenders to build plans for effective defense and assessment of potential risks to a cybersystem. With close to 170k vulnerabilities, manual mapping is not a feasible option. However, automated mapping is challenging due to limited training data, computational intractability, and limitations in computational natural language processing. Tools based on breakthroughs in Transformer-based language models have been demonstrated to classify vulnerabilities with high accuracy. We make three key contributions in this paper: (1) We present a new framework, \VWCBERT, that augments the Transformer-based hierarchical multi-class classification framework of Das et al. (\textsc{V2W-BERT}) with the ability to map weaknesses to exploits. (2) We implement \VWCBERT~ on modern AI accelerator platforms using two data parallel techniques for the pre-training phase and demonstrate nearly linear speedups across NVIDIA and Graphcore accelerator platforms. We observe nearly linear speedups for up to 16 V100 and 8 A100 GPUs, and about 3.4$\times$ speedup for A100 relative to V100 GPUs. We also observe excellent speedups on Graphcore, with $5.7\times$ speedup on 128 IPUs relative to 16 IPUs. Enabled by scaling, we also demonstrate higher accuracy using a larger language model, RoBERTa-Large. We show up to 87\% accuracy for strict and up to 98\% accuracy for relaxed classification. (3) We develop a novel parallel link manager for the link prediction phase and demonstrate up to 21$\times$ speedup with 16 V100 GPUs relative to one V100 GPU, and thus reducing the runtime from 2.5 hours to 10 minutes. We believe that generalizability and scalability of \VWCBERT~ will benefit both the theoretical development and practical deployment of novel cyberdefense solutions and vulnerability classification.

Das, Siddhartha Shankar↗
DOE OSTI2021

Evaluating CUDA Portability with HIPCL and DPCT

HIPCL is expanding the scope of the CUDA portability route from an AMD platform to an OpenCL platform. In the meantime, the Intel DPC++ Compatibility Tool (DPCT) is migrating a CUDA program to a data parallel C++ (DPC++) program. Towards the goal of portability enhancement, we evaluate the performance of the CUDA applications from Rodinia, SHOC, and proxy applications ported using HIPCL and DPCT on Intel GPUs. After profiling the ported programs, we aim to understand their performance gaps, and optimize codes converted by DPCT to improve their performance. The open-source repository for the CUDA, HIP, and DPCT programs will be useful for the development of a translator.

Jin, Zheming↗
DOE OSTI2020

Deep Generative Models that Solve PDEs: Distributed Computing for Training Large Data-Free Models

Recent progress in scientific machine learning (SciML) has opened up the possibility of training novel neural network architectures that solve complex partial differential equations (PDEs). Several (nearly data free) approaches have been recently reported that successfully solve PDEs, with examples including deep feed forward networks, generative networks, and deep encoder-decoder networks. However, practical adoption of these approaches is limited by the difficulty in training these models, especially to make predictions at large output resolutions (≥1024×1024). Here we report on a software framework for data parallel distributed deep learning that resolves the twin challenges of training these large SciML models - training in reasonable time as well as distributing the storage requirements. Our framework provides several out of the box functionality including (a) loss integrity independent of number of processes, (b) synchronized batch normalization, and (c) distributed higher-order optimization methods. We show excellent scalability of this framework on both cloud as well as HPC clusters, and report on the interplay between bandwidth, network topology and bare metal vs cloud. We deploy this approach to train generative models of sizes hitherto not possible, showing that neural PDE solvers can be viably trained for practical applications. We also demonstrate that distributed higher-order optimization methods are 2-3× faster than stochastic gradient-based methods and provide minimal convergence drift with higher batch-size.

PDEs↗
DOE OSTI2023

Scaling Resolution of Gigapixel Whole Slide Images Using Spatial Decomposition on Convolutional Neural Networks

Gigapixel images are prevalent in scientific domains ranging from remote sensing, and satellite imagery to microscopy, etc. However, training a deep learning model at the natural resolution of those images has been a challenge in terms of both, overcoming the resource limit (e.g. HBM memory constraints), as well as scaling up to a large number of GPUs. In this paper, we trained Residual neural Networks (ResNet) on 22,528 x 22,528-pixel size images using a distributed spatial decomposition method on 2,304 GPUs on the Summit Supercomputer. We applied our method on a Whole Slide Imaging (WSI) dataset from The Cancer Genome Atlas (TCGA) database. WSI images can be in the size of 100,000 x 100,000 pixels or even larger, and in this work we studied the effect of image resolution on a classification task, while achieving state-of-the-art AUC scores. Moreover, our approach doesn't need pixel-level labels, since we're avoiding patching from the WSI images completely, while adding the capability of training arbitrary large-size images. This is achieved through a distributed spatial decomposition method, by leveraging the non-block fat-tree interconnect network of the Summit architecture, which enabled GPU-to-GPU direct communication. Finally, detailed performance analysis results are shown, as well as a comparison with a data-parallel approach when possible.

Tsaris, Aristeidis (aris)↗
DOE OSTI2024

KLOKKOS: Automated test generation for performance-portable programs

SAND2024-13830O The KLOKKOS project provides a suite of test generation and validation tools to increase maintainability and portability of Kokkos programs. KLOKKOS, a portmanteau of Klee and Kokkos, is developing an automated test generation for performance-portable programs using Clang/low-level virtual machine (LLVM) and formal methods. KLOKKOS is based on several topics in computer science: the development of symbolic and concolic analysis of LLVM and C++ codes, formal semantics and type theory, C++ templates and language features, LLVM compiler infrastructure, and data parallel primitives. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525

SciDAC↗
DOE OSTI2026

HydraGNN v5.0

HydraGNN v5.0 expands the code base into a more portable, scalable, and flexible framework for scientific graph learning, with particular strength in atomistic machine-learning interatomic potentials and large-scale distributed training. The release adds Fully Sharded Data Parallel (FSDP) support alongside existing DDP and DeepSpeed paths, including FSDP-aware checkpointing and optimizer integration, and introduces a configurable multi-precision training workflow supporting FP32, BF16, and FP64 across GPUs and Intel XPUs. For atomistic modeling, HydraGNN v5.0 strengthens its MLIP capabilities through dynamic graph construction at every forward pass, energy-conserving force prediction via automatic differentiation, and per-atom energy loss formulations, while extending EGNN models to properly handle periodic boundary conditions. The release also broadens model expressiveness through graph-level attribute conditioning, adds new multi-task and model-parallel extensions such as MACE support and encoder/decoder branch optimization, and expands application coverage with integrated examples for datasets including OC25, Nabla2-DFT, QCML, Open Polymers 2026, and OPF. In parallel, HydraGNN v5.0 improves production readiness through performance optimizations for large-scale runs, stratified sampling and linear-regression preprocessing utilities, and tested installation scripts for DOE supercomputers including Frontier, Aurora, Perlmutter, and Andes. Overall, the release advances HydraGNN as a robust software platform for scalable graph neural networks across materials science, chemistry, and scientific machine learning workflows

Lupo Pasini, Massimiliano [Oak Ridge National Labo↗