Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “distributed algorithms”

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 181 records · Page 10
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Adaptive Protection and Validated Models to Enable Deployment of High Penetrations of Solar PV (PV-MOD)

The availability and validation of various PV models in commercial tools differ, with some models not yet thoroughly validated for advanced inverter functionalities and reliable performance under weak system conditions. Many existing models do not fully incorporate new inverter control functions, which can affect system stability. The increasing deployment of solar PV and other inverter-based resources (IBRs), including distributed energy resources (DERs), is influencing the reliable operation of protection schemes in distribution systems and microgrids. Emerging adaptive protection schemes (APS) offer new opportunities for protecting these systems during varying configurations and DER operating conditions, though their demonstration and validation remain limited. Adaptive protection schemes face similar challenges, as they are typically designed for specific configurations. There is a growing need for tools and methodologies to streamline the deployment of adaptive protection for safe and reliable DER integration. The project main objective was to develop and validate high-fidelity generic models of solar PV facilities for stability, protection, EMT, and QSTS analyses. This objective was achieved, and these models can now be integrated into commercial software tools, enabling utilities, vendors, and developers to study high-penetration PV systems more confidently. The project also demonstrated advanced applications of these models, including the design and deployment of adaptive protection schemes in high-penetration field applications and microgrids, supporting grid safety and reliability. Several milestones were reached by the end of the project. A sophisticated inverter test plan was developed, and inverters representative of the North American marketplace were selected. EPRI and NREL tested various inverters, conforming to IEEE standards. Improvements were made to existing generic models of IBR units, IBR plants, and aggregated feeders for various analyses. The first generic electromagnetic transient (EMT) model for a solar PV plant was developed, conforming to IEEE Std 2800™-2022 and validated against laboratory measurements of a 2.2 MVA large-scale battery energy storage system (BESS) inverter. That model was then used to produce reference responses illustrating examples of validated and verified IBR plant models that pass or fail tests for technical minimum capability and performance as specified in the IEEE standard. The developed, tested, and validated generic models can be used for transmission planning, stability assessments, expansion planning, and evaluating potential future IBR interconnection requirements. They can also support interconnection screens and conformity assessments of IBR plants, including solar PV. The project significantly contributed to the ongoing standardization and model-based representation and verification of IBR responses. The project further addressed challenges of common distribution protection schemes with increasing deployment of DER by developing, validating, and demonstrating adaptive protection schemes (APS) that can improve the reliable and safe integration of DER into distribution systems. New APS were designed using improved DER models for three common distribution systems: a radial feeder, a meshed network, and a microgrid. Modeling and hardware-in-the-loop (HIL) testing of the APS were conducted, successfully showing their effectiveness and selectivity. Proof-of-concept field demonstration was achieved for two APS, i.e., one on a radial feeder and another one in a microgrid. Field demonstration could not be achieved for the APS on a meshed network, primarily due apprehension of one utility partner and also due to limited access to the protective algorithms in the network protectors. Guidelines developed from the lessons learned in the project lay out the general process followed in the design, installation, and commissioning of APS for various distribution systems. Distribution utility partners’ apprehension about field demonstration of the new APS were addressed—with varying success—by taking a stepped risk-management approach of modeling of a wide range of sensitivities first, performing in-depth proof-of-concept testing in the laboratory including HIL next, and finally deliberately implementing and commissioning the actual protection equipment and algorithms into parts of—or in parallel operation to—the three real distribution systems. Future work should include pilot projects that further show the acceptable performance of the developed APS before these schemes be rolled out more widely. Inclusion of both utility and original equipment manufacturers (OEMs) in future projects could increase chances of successful field demonstration. Despite challenges in achieving the field demonstration goal of the project for all three APS, the research significantly contributed to the innovation of adaptive protection solutions for scalable and reliable DER integration into distribution systems. This project significantly enhances the understanding of the impact of using appropriate inverter models on distribution and transmission (T&D) systems. By addressing the limitations of existing generic models, the project introduces high-fidelity models for stability, protection, electromagnetic transient (EMT), and quasi-static time series (QSTS) analyses. These models, integrated into commercial software tools, enable utilities, vendors, and developers to confidently study high-penetration PV systems. The project also demonstrates advanced applications, including adaptive protection schemes (APS) for distribution systems and microgrids, ensuring grid safety and reliability. The technical effectiveness and economic feasibility of the methods are evident through the development and validation of sophisticated inverter test plans and the selection of representative inverters. Testing by EPRI and NREL on retail, commercial, and utility-scale inverters, conforming to IEEE standards, underscores the robustness of the models. Improvements to existing generic models for various analyses further enhance their validity and applicability. The project also identifies gaps in common distribution protection schemes and designed new APS using improved DER models, demonstrating their effectiveness through modeling and hardware-in-the-loop (HIL) testing. The project’s benefits to the public are manifold. By advancing the standardization and model-based representation of IBR response, it supports transmission planning, stability assessments, and future IBR interconnection requirements. The generic models can facilitate better communication between transmission planners and developers, supporting expected IBR plant capability and performance. Additionally, the development of APS for radial feeders, meshed networks, and microgrids supports the integration of distributed energy resources (DERs) into distribution systems, enhancing grid reliability and safety. The project’s emphasis on thorough testing and simplicity in design ensures practical and scalable solutions for DER integration.

14 SOLAR ENERGY↗
DOE OSTI2024

A Solution Method for the Filtered Lifting Line Theory

The filtered lifting line theory presents a continuous form of the inviscid momentum equations of flow over a lifting device, such as a wing or rotor blade, using body forces without mathematical singularities. This theory is also consistent with an actuator line representation of a lifting device. In this work, we present a reformulation of the equations in terms of the local flow angle along the line, which allows solving the stand-alone equations using multivariate root-finding algorithms. This approach can be used to obtain a fast, computationally inexpensive solution of the loading distribution along a wing without the need to perform computational fluid dynamic simulations. We study the requirements in terms of resolution in the spanwise direction and establish the criteria for spacing and minimum amount of points required along the blade to obtain converged solutions. The solutions are compared to results from large-eddy simulations, and we observed excellent agreement with less than a percent difference in quantities along the blade between the methods.

17 WIND ENERGY↗
DOE OSTI2025

Communication Lower Bounds and Optimal Algorithms for Symmetric Matrix Computations

In this article, we focus on the communication costs of three symmetric matrix computations: (i) multiplying a matrix with its transpose, known as a symmetric rank-k update (SYRK) (ii) adding the result of the multiplication of a matrix with the transpose of another matrix and the transpose of that result, known as a symmetric rank-2k update (SYR2K) (iii) performing matrix multiplication with a symmetric input matrix (SYMM). All three computations appear in the Level 3 Basic Linear Algebra Subroutines (BLAS) and have wide use in applications involving symmetric matrices. We establish communication lower bounds for these kernels using sequential and distributed-memory parallel computational models, and we show that our bounds are tight by presenting communication-optimal algorithms for each setting. Our lower bound proofs rely on applying a geometric inequality for symmetric computations and analytically solving constrained nonlinear optimization problems. As a result, the symmetric matrix and its corresponding computations are accessed and performed according to a triangular block partitioning scheme in the optimal algorithms.

Al Daas, Hussam [Rutherford Appleton Laboratory, D↗
DOE OSTI2024

Reconstruction of the 4D beam matrix

The widely used transverse parameters characterizing particle beams are the Twiss parameters. These parameters can be measured experimentally but they do not fully characterize the beam since they do not account for possible correlations in particle distribution between two transverse coordinates. These correlations may occur due to uncompensated magnetic field at the cathode or misalignment of focusing quadrupoles in the transport beamline. We test a novel diagnostic for diagnosing full 4D beam matrix which may be used to identify such imperfections. The diagnostic is based on transporting the beam through the beamline which includes a quadrupole and a skew quadrupole magnets and measuring the resulting 2D beam distribution at the screen downstream. Such a measurement can be viewed as measuring a 2D projection of the 4D distribution. Different settings of the quads provide measurements of different slices of the phase space. The reconstruction of the original beam matrix from a number of measurements is done using machine learning algorithm, which provides a fast and reliable way of reconstruction for an arbitrary configuration of the scanning beamline. In August 2024, we set up the diagnostic beamline to perform a quadrupole scan of the beam. The setup includes a skew quadrupole, a regular quadrupole, and a screen. The images on the screen were post-processed to remove experimental artifacts and enhance contrast by eliminating background noise outside the core of the distribution=. The rms parameters of the distribution were then calculated and used as inputs for the reconstruction algorithm. This algorithm attempts to determine the initial beam matrix that produces expected images on the screen closely matching the observed images across all quadrupole settings. The algorithm found a solution in which the expected rms parameters closely align with the observations. Validation of the results is planned for FY25.

43 PARTICLE ACCELERATORS↗
DOE OSTI2025

A scoping study of far-SOL main-wall protection limiters for steady-state operation of compact pilot plant tokamaks

We present a novel method for handling steady-state heat fluxes incident on the main wall of pilot plant-scale magnetic fusion devices, based on the utilization of protection limiters in the far scrape-off layer (SOL). This method helps avoid large plasma-wall gaps, without excessively compromising blanket performance. We present an optimization algorithm for determining the appropriate size and scale of these protection limiters given (1) probability distributions of SOL plasma parameters and (2) assumed risk tolerance. As part of this optimization, we have developed an analytic description of parallel heat fluxes across limiter shadows, and an objective cost function (the ‘Far-SOL Marginal Cost’) to quantify the impact that different main-wall thermal management design choices have on reactor capital cost. Applying the model to a midscale fusion pilot plant concept shows that making use of far-SOL protection limiters can reduce capital costs on the order of $500 M, relative to naively increasing the plasma-wall gap. Our analysis demonstrates that the far-SOL power decay length is the highest-leverage plasma assumption for thermal loading of the first wall, and the primary cost driver for main wall thermal management. The relative cost efficiency of protection limiters increases as assumptions on the far-SOL heat flux become more pessimistic. The concepts described in this paper motivate the further development of far-SOL protection limiters as part of larger efforts to design economical core-edge-wall compatible solutions for a fusion pilot plant.

Design under uncertainty↗
DOE OSTI2026

Sampling two-dimensional isometric tensor network states

Sampling a quantum system’s underlying probability distributions is an important computational task, e.g., for quantum advantage experiments and quantum Monte Carlo algorithms. Tensor networks are an invaluable tool for efficiently representing states of large quantum systems with limited entanglement. Algorithms for sampling one-dimensional (1D) tensor networks are well-established and utilized in several 1D tensor network methods. In this paper we introduce two novel sampling algorithms for two-dimensional (2D) isometric tensor network states (isoTNS) that generalize existing 1D tensor network sampling algorithms. Our first proposed algorithm performs independent sampling and yields a single configuration together with its associated probability. The second algorithm employs a greedy search strategy to identify high-probability configurations and their corresponding probabilities. Numerical results demonstrate the effectiveness of these algorithms across quantum states with varying entanglement and system size.

Dumitrescu, Eugene [ORNL] (ORCID:0000000158519567)↗
DOE OSTI2025

Single-shot in-line x-ray phase-contrast imaging of void-shockwave interactions in fusion energy materials

Recent breakthroughs in nuclear fusion, specifically the report of reactions exceeding scientific breakeven at the National Ignition Facility (NIF), highlight the potential of inertial fusion energy (IFE) as a sustainable and virtually limitless energy source. However, further progress in IFE requires characterization of defects in ablator materials and how they affect fuel capsule compression. Voids within the ablator can degrade energy yield, but their impact on the density distribution has primarily been studied through simulations, with limited high-resolution experimental validation. To address this, we used the x-ray free-electron laser (XFEL) at the matter in extreme conditions (MECs) instrument at the Linac coherent light source (LCLS) to capture 2D x-ray phase-contrast (XPC) images of a void-bearing sample with a composition similar to inertial confinement fusion (ICF) ablators. By driving a compressive shockwave through the sample using MEC's long-pulse laser system, we analyzed how voids influence shockwave propagation and density distribution during compression. To quantify this impact, we extracted phase information using two phase retrieval algorithms. First, we applied the contrast transfer function (CTF) method, paired with Tikhonov regularization and a fast optimization approach to generate an initial phase estimate. We then refined the result using a projected gradient descent (PGD) method that works directly with the sample's refractive index. Comparing these results with radiation adaptive grid Eulerian (xRAGE) radiation hydrodynamic simulations enables identification of model validation needs or improvements. By calculating phase maps in situ, it becomes possible to reconstruct areal density maps, improving understanding of laser-capsule interactions and advancing IFE research.

Hodge, D. S. [Colorado State Univ., Fort Collins, ↗
DOE OSTI2024

FuseIM: Fusing Probabilistic Traversals for Influence Maximization on Exascale Systems

Probabilistic breadth-first traversals (BPTs) are used in many network science and graph machine learning applications. In this paper, we are motivated by the application of BPTs in stochastic diffusion-based graph problems such as influence maximization. These applications heavily rely on BPTs to implement a Monte-Carlo sampling step for their approximations. Given the large sampling complexity, stochasticity of the diffusion process, and the inherent irregularity in real-world graph topologies, efficiently parallelizing these BPTs remains significantly challenging. In this paper, we present a new algorithm to fuse massive number of concurrently executing BPTs with random starts on the input graph. Our algorithm is designed to fuse BPTs by combining separate traversals into a unified frontier on distributed multi-GPU systems. To show the general applicability of the fused BPT technique, we have incorporated it into two state-of-the-art influence maximization parallel implementations (gIM and Ripples). Our experiments on up to 4K nodes of the OLCF Frontier supercomputer (32,768 GPUs and 196K CPU cores) show strong scaling behavior, and that fused BPTs can improve the performance of these implementations up to 34x (for gIM) and ~360x (for Ripples).

Neff, Reece W.↗
DOE OSTI2024

On optimizing the sensor spacing for pressure measurements on wind turbine airfoils

This research article presents a robust approach to optimizing the layout of pressure sensors around an airfoil. A genetic algorithm and a sequential quadratic programming algorithm are employed to derive a sensor layout best suited to represent the expected pressure distribution and, thus, the lift force. The fact that both optimization routines converge to almost identical sensor layouts suggests that an optimum exists and is reached. By comparing against a cosine-spaced sensor layout, it is demonstrated that the underlying pressure distribution can be captured more accurately with the presented layout optimization approach. Conversely, a 39 %–55 % reduction in the number of sensors compared to cosine spacing is achievable without loss in lift prediction accuracy. Given these benefits, an optimized sensor layout improves the data quality, reduces unnecessary equipment and saves cost in experimental setups. While the optimization routine is demonstrated based on the generic example of the IEA 15 MW reference wind turbine, it is suitable for a wide range of applications requiring pressure measurements around airfoils.

17 WIND ENERGY↗
DOE OSTI2026

Distributed Quantum-Enhanced Optimization: A Topographical Preconditioning Approach for High-Dimensional Search

Optimization problems become fundamentally challenging as the number of variables increases. Because the volume of the search space grows exponentially, classical algorithms frequently fail to locate the global minimum of non-convex functions. While quantum optimization offers a potential alternative, mapping continuous problems onto near-term quantum hardware introduces severe scaling limits and barren plateaus. To bridge this gap, we propose the Distributed Quantum-Enhanced Optimization (D-QEO) framework. Instead of forcing the quantum processor to find the exact minimum, we use it simply as a topographical preconditioner. The QPU maps the landscape to locate the most promising basin of attraction, generating high-quality seed points for a classical GPU-accelerated solver to refine. To make this approach viable for utility-scale problems, we exploit the mathematical structure of separable functions. This allows us to cut a 50-qubit (i.e., $2^{50}$) global search space into independent and manageable sub-spaces using 5-qubit subcircuits. By executing these fragments concurrently with CUDA-Q, we completely bypass the overhead of cross-register entanglement and classical tensor knitting for separable functions. Benchmarks on the 10-dimensional Rastrigin and Ackley functions show that D-QEO prevents the exponential failure rates observed in purely classical algorithms. Furthermore, this quantum warm-start significantly reduces the number of classical BFGS iterations required to converge, providing a highly practical blueprint for utilizing near-term quantum resources in complex global search.

Soos, Dominik [Old Dominion U.]↗
DOE OSTI2025

End-to-end deep learning pipeline for real-time Bragg peak segmentation: from training to large-scale deployment

X-ray crystallography reconstruction, which transforms discrete X-ray diffraction patterns into three-dimensional molecular structures, relies critically on accurate Bragg peak finding for structure determination. As X-ray free electron laser (XFEL) facilities advance toward MHz data rates (1 million images per second), traditional peak finding algorithms that require manual parameter tuning or exhaustive grid searches across multiple experiments become increasingly impractical. While deep learning approaches offer promising solutions, their deployment in high-throughput environments presents significant challenges in automated dataset labeling, model scalability, edge deployment efficiency, and distributed inference capabilities. We present an end-to-end deep learning pipeline with three key components: (1) a data engine that combines traditional algorithms with our peak matching algorithm to generate high-quality training data at scale, (2) a modular architecture that scales from a few million to hundreds of million parameters, enabling us to train large expert-level models offline while deploying smaller, distilled models at the edge, and (3) a decoupled producer-consumer architecture that separates specialized data source layer from model inference, enabling flexible deployment across diverse computing environments. Using this integrated approach, our pipeline achieves accuracy comparable to traditional methods tuned by human experts while eliminating the need for experiment-specific parameter tuning. Although current throughput requires optimization for MHz facilities, our system's scalable architecture and demonstrated model compression capabilities provide a foundation for future high-throughput XFEL deployments.

Wang, Cong↗
DOE OSTI2026

Noise-aware optimization in nominally identical manufacturing and measuring systems for high-throughput parallel workflows

Device-to-device variability in experimental noise critically impacts reproducibility, especially in automated, high-throughput systems like additive manufacturing farms. While manageable in small labs, such variability can escalate into serious risks at larger scales, such as architectural 3D printing, where noise may cause structural or economic failures. This contribution presents a noise-aware decision-making algorithm that quantifies and models device-specific noise profiles to manage variability adaptively. It uses distributional analysis and pairwise divergence metrics with clustering to choose between single-device and robust multi-device Bayesian optimization strategies. Unlike conventional methods that assume homogeneous devices or enforce generic robustness, the proposed framework explicitly determines whether shared optimization across devices is appropriate based on the degree of inter-device noise heterogeneity. This enables improved performance, reproducibility, and efficiency. An experimental case study involving three nominally identical 3D printers (same brand, model, and close serial numbers) demonstrates reduced redundancy, lower resource usage, and improved reliability, along with improved convergence stability and solution quality through the selection of the appropriate optimization strategy based on the degree of inter-device noise heterogeneity. Overall, this framework establishes a general approach for precision- and resource-aware optimization in scalable, automated experimental platforms, demonstrated here on a representative multi-device 3D printing case study.

Schenk, Christina↗
DOE OSTI2024

Jet radius dependence of dijet momentum balance and suppression in Pb + Pb collisions at 5.02 TeV with the ATLAS detector

This paper describes a measurement of the jet radius dependence of the dijet momentum balance between leading back-to-back jets in 1.72 nb -1 of Pb + Pb collisions collected in 2018 and 255 pb -1 of pp collisions collected in 2017 by the ATLAS detector at the LHC. Both datasets were collected at $\sqrt{s_{NN}}$ = 5.02 TeV. Jets are reconstructed using the anti-k t algorithm with jet radius parameters R = 0.2, 0.3, 0.4, 0.5, and 0.6. The dijet momentum balance distributions are constructed for leading jets with transverse momentum p T from 100 to 562 GeV for R = 0.2, 0.3, and 0.4 jets, and from 158 to 562 GeV for R = 0.5 and 0.6 jets. The absolutely normalized dijet momentum balance distributions are constructed to compare measurements of the dijet yields in Pb + Pb collisions directly to the dijet cross sections in pp collisions. For all jet radii considered here, there is a suppression of more balanced dijets in Pb + Pb collisions compared with pp collisions, while for more imbalanced dijets there is an enhancement. There is a jet radius dependence to the dijet yields, being stronger for more imbalanced dijets than for more balanced dijets. Additionally, jet pair nuclear modification factors are measured. The subleading jet yields are found to be more suppressed than leading jet yields in dijets. A jet radius dependence of the pair nuclear modification factors is observed, with the suppression decreasing with increasing jet radius. These measurements provide new constraints on jet quenching scenarios in the quark-gluon plasma.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2025

Single-Shot Reconstruction of Electron Beam Longitudinal Phase Space in a Laser Wakefield Accelerator

We report on a single-shot longitudinal phase-space reconstruction diagnostic for electron beams in a laser wakefield accelerator via the experimental observation of distinct periodic modulations in the angularly resolved spectra. Such modulated angular spectra arise as a result of the direct interaction between the ultrarelativistic electron beam and the laser driver in the presence of the wakefield. A constrained theoretical model for the coupled oscillator, assisted by a genetic algorithm, can recreate the experimental electron spectra and, thus, fully reconstructs the longitudinal phase-space distribution of the electron beam with a temporal resolution of approximately 1.3 fs. In particular, it reveals the slice energy spread of the electron beam, which is important to measure for applications such as x-ray free electron lasers. In our experiment, the root-mean-square energy spread retrieved is bounded at 9.9 MeV, corresponding to a 0.9%–3.0% relative spread, despite the overall GeV energy beam having approximately 100% relative energy spread.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2025

Characterization and Quantification of Radiation-Induced Clusters/Precipitates in RPV Steels Using STEM-EDS and Machine Learning

Over the operational lifespan of a nuclear reactor, reactor pressure vessel (RPV) steels are subjected to significant neutron irradiation, resulting in complex microstructural changes and the consequent degradation of mechanical properties. Various physically motivated correlation models have been developed to predict neutron irradiation-induced embrittlement of RPVs under different irradiation conditions. However, the efficient and accurate characterizations and quantification of radiation-induced clusters in RPVs are still challenging, which will affect the precision of the predictive models for embrittlement of RPV components. In the DOE Visiting Faculty Program (VFP) research work at Oak Ridge National Lab (ORNL), I integrate machine learning to aid Scanning Transmission Electron Microscopy – Energy Dispersive X-ray Spectroscopy (STEM-EDS) analyses, which improve the characterization and quantification of radiation-induced clusters in RPV steels, thereby enabling more accurate predictions of material behavior under irradiation. The surveillance base- and welded- RPV steels were annealed at various temperatures of 340 °C, 450 °C and 500 °C for up to 168 hours, respectively. Afterwards, I have characterized radiation-induced clusters using advanced STEM-EDS techniques and subsequently applying machine learning algorithms to analyze and refine STEM-EDS datasets, enhancing the quantification of clusters compositions and distributions. In the end, an efficient workflow for integrating STEM-EDS data analysis with machine learning to address challenges including noise reduction has been developed. The completion of this VFP work will support bridge critical gaps in the accurate quantification of radiation-induced clusters in RPV steels using STEM-EDS and support the development of more precise models for predicting RPV embrittlement in the Light Water Reactor Sustainability program supported by Department of Energy and enhancing the collaboration between ORNL and Alred University. The outcome of the VFP project will leverage a few research papers submission to peer-reviewed journals in the relevant scientific field and a few oral presentations at national and international conferences.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2024

CACTI CSAPR2 Taranis Retrievals

Taranis is an end-to-end processing chain for radar data written in Python with C extension for computation performance. Features include: masking for quality control, specific differential phase (Kdp), attenuation correction for reflectivity factor (Z) and differential reflectivity (Zdr) in rain, and additional geophysical retrievals. Retrievals are mostly drawn from literature or open-source software when appropriate, and have been tested, tuned, and modified to work with one another cohesively rather than using isolated off-the-shelf algorithms. Incorporated algorithms include hydrometeor (echo) identification, rain water content, raindrop mass-weighted mean diameter (gamma size distribution assumption), and rainfall rate (QPE). Taranis data sets exist for CSAPR2 PPI, HSRHI, and sector RHI scans. Cartesian-gridded data sets were also produced as well as a near-surface rain rate retrieval. More details can be found in the README.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Unpaired image translation to mitigate domain shift in liquid argon time projection chamber detector responses

Deep learning algorithms often are developed and trained on a training dataset and deployed on test datasets. Any systematic difference between the training and a test dataset may severely degrade the final algorithm performance on the test dataset—what is known as the domain shift problem . This issue is prevalent in many scientific domains where algorithms are trained on simulated data but applied to real-world datasets. Typically, the domain shift problem is solved through various domain adaptation (DA) methods. However, these methods are often tailored for a specific downstream task, such as classification or semantic segmentation, and may not easily generalize to different tasks. This work explores the feasibility of using an alternative way to solve the domain shift problem that is not specific to any downstream algorithm. The proposed approach relies on modern Unpaired Image-to-Image (UI2I) translation techniques, designed to find translations between different image domains in a fully unsupervised fashion. In this study, the approach is applied to a domain shift problem commonly encountered in Liquid Argon Time Projection Chamber (LArTPC) detector research when seeking a way to translate samples between two differently distributed LArTPC detector datasets deterministically. This translation allows for mapping real-world data into the simulated data domain where the downstream algorithms can be run with much less domain-shift-related performance degradation. Conversely, using the translation from the simulated data to a real-world domain can increase the realism of the simulated dataset and reduce the magnitude of any systematic uncertainties. To evaluate the quality of the translations, we use both pixel-wise metrics and a downstream task to measure the effectiveness of UI2I methods for mitigating the domain shift problem. We adapted several popular UI2I translation algorithms to work on scientific data and demonstrated the viability of these techniques for solving the domain shift problem with LArTPC detector data. To facilitate further development of DA techniques for scientific datasets, the ‘Simple Liquid-Argon Track Samples’ dataset used in this study is also published.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

CG-Kit: Code Generation Toolkit for performant and maintainable variants of source code applied to Flash-X hydrodynamics simulations

CG-Kit is a new Code Generation tool-Kit that we have developed as a part of the solution for portability and maintainability for multiphysics computing applications. The development of CG-Kit is rooted in the urgent need created by the shifting landscape of high-performance computing platforms and the algorithmic complexities of a particular large-scale multiphysics application: Flash-X. To efficiently use computing resources on a heterogeneous node, an application must have a map of computation to resources and a mechanism to move the data and computation to the resources according to the map. Most existing performance portability solutions are focussed on abstracting the expression of computations so that a unified source code can be specialized to run on different resources. However, such an approach is insufficient for a code like Flash-X, which has a multitude of code components that can be assembled in various permutations and combinations to form different instances of applications. Similar challenges apply to any code that has composability, where a single specified way of apportioning work among devices may not be optimal. Additionally, use cases arise where the optimal control flow of computation may differ for different devices while the underlying numerics remain identical. This combination leads to unique challenges including handling an existing large code base in Fortran and/or C/C++, subdivision of code into a great variety of units supporting a wide range of physics and numerical methods, different parallelization techniques for distributed and shared memory systems and accelerator devices, and heterogeneity of computing platforms requiring coexisting variants of parallel algorithms. All of these challenges demand that scientific software developers apply existing knowledge about domain applications, algorithms, and computing platforms to determine custom abstractions and granularity for code generation. There is a critical lack of tools to tackle those problems. CG-Kit is designed to fill this gap by providing a user with the ability to express their desired control flow and computation-to-resource map in the form a pseudocode-like recipe. It consists of standalone tools that can be combined into highly specific and, we argue, highly effective portability and maintainability toolchains. Here we present the design of our new tools: parametrized source trees, control flow graphs, and recipes. The tools are implemented in Python. They are agnostic to the programming language of the source code targeted for code generation. In conclusion, we demonstrate the capabilities of the toolkit with two examples, first, multithreaded variants of the basic AXPY operation, and second, variants of parallel algorithms within a hydrodynamics solver, called Spark, from Flash-X that operates on block-structured adaptive meshes.

Algorithmic portability↗