SEARCH · Engineering Papers

Results for “scalable 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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 397 records · Page 22

All sources NASA NTRS DOE OSTI

DOE OSTI2025

Scalable Hybrid Learning Techniques for Scientific Data Compression

Data compression is becoming critical for storing scientific data because many scientific applications need to store large amounts of data and post process this data for scientific discovery. Unlike image and video compression algorithms that limit errors to primary data (PD), scientists require compression techniques that accurately preserve derived quantities of interest (QoIs). Here, this article presents a physics-informed compression technique implemented as an end-to-end, scalable, GPU-based pipeline for data compression that addresses this requirement. Our hybrid compression technique combines machine learning techniques and standard compression methods. Specifically, we combine an autoencoder, an error-bounded lossy compressor to provide guarantees on raw data error, and a constraint satisfaction post-processing step to preserve the QoIs within a minimal error (generally less than floating point error). The effectiveness of the data compression pipeline is demonstrated by compressing nuclear fusion simulation data generated by a large-scale fusion code, XGC, which produces hundreds of terabytes of data in a single day. Our approach works within the ADIOS framework and results in compression by a factor of more than 150 while requiring only a few percent of the computational resources necessary for generating the data, making the overall approach highly effective for practical scenarios.

ITER↗

DOE OSTI2024

Mid-IR UAV-based sensing platform with deep learning to Identify and Quantify Gaseous Emission in Gas Flares

This report details the development and evaluation of a Mid-Infrared (Mid-IR) Unmanned Aerial Vehicle (UAV)-based sensing platform integrated with deep learning algorithms for the identification and quantification of gaseous emissions in gas flares. The project, spearheaded by Omega Optics, Inc., aimed to address environmental monitoring challenges by leveraging advanced photonic technologies and autonomous UAV operations. The research focused on designing, optimizing, and fabricating photonic crystal waveguides and grating couplers to enhance the sensitivity and accuracy of gas detection. A comprehensive drone-based system was developed, featuring a miniaturized sensor, GPS module, and microcontroller communication network for real-time gas concentration monitoring. The system's adaptive sampling algorithm, implemented using the Robot Operating System (ROS), enables autonomous detection and localization of gas emission sources. Preliminary results demonstrate the platform's capability to detect and monitor gas emissions with high precision, cost-effectiveness, and scalability. Future work will expand upon this foundation by introducing 3D wind model-based learning for dynamic environmental conditions and further enhancing the user interface and data processing algorithms to support broader environmental monitoring applications. Overall, this project represents a significant step forward in UAV-based environmental sensing technologies, offering robust solutions for detecting and mitigating the impacts of gaseous emissions on public health and safety.

47 OTHER INSTRUMENTATION↗

DOE OSTI2020

An open source analysis framework for large-scale building energy modeling

Full integration of building energy modelling into the design and retrofit process has long been a goal of building scientists and practitioners. However, significant barriers still exist. Among them are the lack of available: (1) configurable technology stacks for performing both small- and large-scale analyses, (2) different classes of algorithms compatible with common design workflows, and (3) analysis tools for effectively visualizing large-scale simulation results. This article discusses the OpenStudio® Analysis Framework: a scalable analysis framework for building energy modelling that was developed to overcome the three barriers listed above. The framework is open-source and scalable to facilitate wider adoption and has a clearly defined application programming interface upon which other applications can be built. It runs on high-performance computing systems, within cloud infrastructure, and on laptops, and uses a common workflow to enable different classes of algorithms. Lessons learned from previous development efforts are also discussed.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2021

Efficient loading of reduced data ensembles produced at ORNL SNS/HFIR neutron time-of-flight facilities

We present algorithmic improvements to the loading operations of certain reduced data ensembles produced from neutron scattering experiments at Oak Ridge National Laboratory (ORNL) facilities. Ensembles from multiple measurements are required to cover a wide range of the phase space of a sample material of interest. They are stored using the standard NeXus schema on individual HDF5 files. This makes it a scalability challenge, as the number of experiments stored increases in a single ensemble file. The present work follows up on our previous efforts on data management algorithms, to address identified input output (I/O) bottlenecks in Mantid, an open-source data analysis framework used across several neutron science facilities around the world. We reuse an in-memory binary-tree metadata index that resembles data access patterns, to provide a scalable search and extraction mechanism. In addition, several memory operations are refactored and optimized for the current common use cases, ranging most frequently from 10 to 180, and up to 360 separate measurement configurations. Results from this work show consistent speed ups in wall-clock time on the Mantid LoadMD routine, ranging from 19% to 23% on average, on ORNL production computing systems. The latter depends on the complexity of the targeted instrument-specific data and the system I/O and compute variability for the shared computational resources available to users of ORNL’s Spallation Neutron Source (SNS) and the High Flux Isotope Reactor (HFIR) instruments. Nevertheless, we continue to highlight the need for more research to address reduction challenges as experimental data volumes, user time and processing costs increase.

Godoy, William↗

NASA NTRS2001

Simple, Scalable, Script-Based Science Processor (S4P)

The development and deployment of data processing systems to process Earth Observing System (EOS) data has proven to be costly and prone to technical and schedule risk. Integration of science algorithms into a robust operational system has been difficult. The core processing system, based on commercial tools, has demonstrated limitations at the rates needed to produce the several terabytes per day for EOS, primarily due to job management overhead. This has motivated an evolution in the EOS Data Information System toward a more distributed one incorporating Science Investigator-led Processing Systems (SIPS). As part of this evolution, the Goddard Earth Sciences Distributed Active Archive Center (GES DAAC) has developed a simplified processing system to accommodate the increased load expected with the advent of reprocessing and launch of a second satellite. This system, the Simple, Scalable, Script-based Science Processor (S42) may also serve as a resource for future SIPS. The current EOSDIS Core System was designed to be general, resulting in a large, complex mix of commercial and custom software. In contrast, many simpler systems, such as the EROS Data Center AVHRR IKM system, rely on a simple directory structure to drive processing, with directories representing different stages of production. The system passes input data to a directory, and the output data is placed in a "downstream" directory. The GES DAAC's Simple Scalable Script-based Science Processing System is based on the latter concept, but with modifications to allow varied science algorithms and improve portability. It uses a factory assembly-line paradigm: when work orders arrive at a station, an executable is run, and output work orders are sent to downstream stations. The stations are implemented as UNIX directories, while work orders are simple ASCII files. The core S4P infrastructure consists of a Perl program called stationmaster, which detects newly arrived work orders and forks a job to run the appropriate executable (registered in a configuration file for that station). Although S4P is written in Perl, the executables associated with a station can be any program that can be run from the command line, i.e., non-interactively. An S4P instance is typically monitored using a simple Graphical User Interface. However, the reliance of S4P on UNIX files and directories also allows visibility into the state of stations and jobs using standard operating system commands, permitting remote monitor/control over low-bandwidth connections. S4P is being used as the foundation for several small- to medium-size systems for data mining, on-demand subsetting, processing of direct broadcast Moderate Resolution Imaging Spectroradiometer (MODIS) data, and Quick-Response MODIS processing. It has also been used to implement a large-scale system to process MODIS Level 1 and Level 2 Standard Products, which will ultimately process close to 2 TB/day.

Lynnes, Christopher↗

DOE OSTI2021

SV-Sim: Scalable PGAS-based State Vector Simulation of Quantum Circuits

High-performance quantum circuit simulation in a classic HPC is still imperative in the NISQ era. Observing that the major obstacle of scalable state-vector quantum simulation arises from the massively fine-grained irregular data-exchange with remote nodes, in this paper we present SV-Sim to apply the emerging PGAS-based communication models (i.e., direct peer access for intra-node CPUs/GPUs and SHMEM for inter-node CPU/GPU clusters) for efficient scalable quantum circuit simulation. Through an orchestrated device functional pointer design, SV-Sim is able to abstract the quantum gate sets across various heterogeneous backends, including IBM/Intel/AMD CPUs, NVIDIA /AMD GPUs, and Intel MIC, in a unified framework, but still asserting outstanding performance and tractable interface to higher-level quantum programming environments, such as IBM Qiskit, Microsoft Q\# and Google Cirq. Circumventing the disability of polymorphism in GPUs and leveraging the device-initiated one-sided communication, SV-Sim can process dynamically synthesized quantum circuit in a single GPU/CPU kernel without the need of expensive JIT or runtime branching, significantly improving the performance and simplifying the programming complexity for the emerging variational quantum algorithms. Evaluations on NVIDIA A100-DGX-1, V100-DGX-2, AMD MI100, ALCF Theta, and OLCF Summit HPCs show that SV-Sim can delivery scalable performance on various state-of-the-art HPC platforms, offering a useful tool for quantum algorithm validation and verification.

Li, Ang↗

DOE OSTI2025

Track reconstruction as a service for collider physics

Optimizing charged-particle track reconstruction algorithms is crucial for efficient event reconstruction in Large Hadron Collider (LHC) experiments due to their significant computational demands. Existing track reconstruction algorithms have been adapted to run on massively parallel coprocessors, such as graphics processing units (GPUs), to reduce processing time. Nevertheless, challenges remain in fully harnessing the computational capacity of coprocessors in a scalable and non-disruptive manner. This paper proposes an inference-as-a-service approach for particle tracking in high energy physics experiments. To evaluate the efficacy of this approach, two distinct tracking algorithms are tested: Patatrack, a rule-based algorithm, and Exa.TrkX, a machine learning-based algorithm. The as-a-service implementations show enhanced GPU utilization and can process requests from multiple CPU cores concurrently without increasing per-request latency. The impact of data transfer is minimal and insignificant compared to running on local coprocessors. This approach greatly improves the computational efficiency of charged particle tracking, providing a solution to the computing challenges anticipated in the High-Luminosity LHC era.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2023

Error statistics and scalability of quantum error mitigation formulas

Quantum computing promises advantages over classical computing in many problems. Nevertheless, noise in quantum devices prevents most quantum algorithms from achieving the quantum advantage. Quantum error mitigation provides a variety of protocols to handle such noise using minimal qubit resources. While some of those protocols have been implemented in experiments for a few qubits, it remains unclear whether error mitigation will be effective in quantum circuits with tens to hundreds of qubits. In this paper, we apply statistics principles to quantum error mitigation and analyse the scaling behaviour of its intrinsic error. We find that the error increases linearly O(ϵN) with the gate number N before mitigation and sublinearly O(ϵ'N γ ) after mitigation, where γ ≈ 0.5, ϵ is the error rate of a quantum gate, and ϵ' is a protocol-dependent factor. The $\sqrt{N}$ scaling is a consequence of the law of large numbers, and it indicates that error mitigation can suppress the error by a larger factor in larger circuits. We propose the importance Clifford sampling as a key technique for error mitigation in large circuits to obtain this result.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS2020

Pterodactyl: The Development and Performance of Guidance Algorithms for a Mechanically Deployed Entry Vehicle

Pterodactyl is a NASA Space Technology Mission Directorate (STMD) project focused on developing a design capabilityfor optimal, scalable, Guidance and Control (G&C) solutions that enable precision targeting for Deployable Entry Vehicles(DEVs). This feasibility study is unique in that it focuses on the rapid integration of targeting performance analysis withstructural &packaging analysis, which is especially challenging for new vehicle and mission designs. This paper will detailthe guidance development and trajectory design process for a lunar return mission, selected to stress the vehicle designsand encourage future scalability. For the five G&C configurations considered, the Fully Numerical Predictor-CorrectorEntry Guidance (FNPEG) was selected for configurations requiring bank angle guidance and FNPEG with Uncoupled

Johnson, Breanna↗

DOE OSTI2020

Scalable All-pairs Shortest Paths for Huge Graphs on Multi-GPU Clusters

We present an optimized Floyd-Warshall (Floyd-Warshall) algorithm that computes the All-pairs shortest path (APSP) for GPU accelerated clusters. The Floyd-Warshall algorithm due to its structural similarities to matrix-multiplication is well suited for highly parallel GPU architectures. To achieve high parallel efficiency, we address two key algorithmic challenges: reducing high communication overhead and addressing limited GPU memory. To reduce high communication costs, we redesign the parallel (a) to expose more parallelism, (b) aggressively overlap communication and computation with pipelined and asynchronous scheduling of operations, and (c) tailored MPI-collective. To cope with limited GPU memory, we employ an offload model, where the data resides on the host and is transferred to GPU on-demand. The proposed optimizations are supported with detailed performance models for tuning. Our optimized parallel Floyd-Warshall implementation is up to 5x faster than a strong baseline and achieves 8.1 PetaFLOPS/sec on 256~nodes of the Summit supercomputer at Oak Ridge National Laboratory. This performance represents 70% of the theoretical peak and 80% parallel efficiency. The offload algorithm can handle 2.5x larger graphs with a 20% increase in overall running time.

Sao, Piyush↗

DOE OSTI2021

A Scalable Meter Placement Method for Distribution System State Estimation

This paper studies the optimal meter placement problem for distribution system state estimation given limited measurement resources. We formulate the problem as a mixed integer semi-definite programming that minimizes the worst case estimation errors over a set of operating points. To solve the problem, we first relax the problem as a convex optimization problem. Motivated by the lack of scalability of existing solvers, we next leverage the special structure of the cost function and propose an algorithm based on barrier method that solves the problem with significantly better numerical performance. The proposed method has been validated on the IEEE 13-bus, IEEE 123-bus, and IEEE 8,500-bus feeders.

barrier method↗

DOE OSTI2023

Verifying the Computational Integrity of Power Grid Controls with Zero-Knowledge Proof

The control of future power grids is migrating from a centralized to a distributed/decentralized scheme to enable a massive penetration of distributed energy resources with rising dependence on communication infrastructure. A common assumption made for most existing distributed/decentralized controllers is that local controllers would faithfully follow the designated controller dynamics based on the data received from communication channels. However, with increased probability of cyberattacks on Operational Technology infrastructure, such an assumption could be risky because proper execution of the controller dynamics is then built on trust in secure communication and computation. In this work, we leverage a cryptography technology known as zero-knowledge scalable transparent arguments of knowledge (zk-STARK) to verify the computational integrity of power grid control algorithms, with projected linear dynamics-based control schemes as the initial proof-of-concept test case. The method presented here converts the cybersecurity challenge of data integrity for grid control into a subset of computational integrity.

computational integrity↗

DOE OSTI2026

Accelerating Bilevel Optimization With Hierarchical Many-Threaded Parallel Differential Evolution

Bilevel optimization is encountered in many relevant real-world applications. The main feature of this type of problem is that an upper-level optimization problem is constrained by a nested lower-level optimization problem. Because of this nested structure, bilevel problems (BLPs) are usually computationally expensive to solve. Differential evolution (DE) has demonstrated promising results in solving BLPs of relatively small scales. As the problem scale increases, the decision space becomes intrinsically larger, requiring a growing number of function evaluations for the method to work properly. In this context, heavy parallelization and high-performance computing techniques are indispensable to enable the resolution of more complex and challenging optimization problems. Hence, we propose a hierarchical many-threaded parallel DE approach for BLPs, where both levels are parallelized. The computational experiments demonstrate that the parallel implementation achieved runtime speeds ranging from 44 to 2559 times faster than the sequential version on a well-known scalable SMD benchmark test problem when executed on an NVIDIA A100 GPU. The findings indicate that the algorithm’s convergence is strongly influenced by the number of both upper- and lower-level generations. Moreover, the success of experiments with large-scale problems is closely linked to the choice of small population sizes.

Dufek, Amanda S↗

DOE OSTI2025

A Class of Sparse Johnson–Lindenstrauss Transforms and Analysis of their Extreme Singular Values

The Johnson–Lindenstrauss (JL) lemma is a powerful tool for dimensionality reduction in modern algorithm design. The lemma states that any set of high-dimensional points in a Euclidean space can be projected into lower dimensions while approximately preserving pairwise Euclidean distances. Random matrices satisfying this lemma are called JL transforms (JLTs). Inspired by existing $s$-hashing JLTs with exactly $s$ nonzero elements on each column, the present work introduces an ensemble of sparse matrices encompassing so-called $s$-hashing-like matrices whose expected number of nonzero elements on each column is $s$. The independence of the sub-Gaussian entries of these matrices and the knowledge of their exact distribution play an important role in their analyses. Using properties of independent sub-Gaussian random variables, these matrices are demonstrated to be JLTs, and their smallest nontrivial singular values and largest singular values are estimated nonasymptotically using a technique from geometric functional analysis. As the dimensions of the matrix grow to infinity, these singular values are proved to converge almost surely to fixed quantities (by using the universal Bai–Yin law) and in distribution to the Gaussian orthogonal ensemble Tracy–Widom law after proper rescalings. Understanding the behaviors of extreme singular values is important in general because they are often used to define a measure of stability of matrix algorithms. For example, JLTs were recently used in derivative-free optimization algorithmic frameworks to select random subspaces in which are constructed random models or poll directions to achieve scalability, and hence estimating their smallest singular value in particular helps determine the dimension of these subspaces.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Open Reproducible Electron Microscopy Data Analysis

Electron microscopy (EM) is a cornerstone technique in the materials and biological sciences capable of imaging structures at nano- to atomic-scale resolution. Advances in technologies mean that one acquires datasets at increasing data rates and sizes. These advancements present enormous opportunities for researchers to understand complex systems. However, processing the resulting large-scale, complex data in a reproducible and shareable way is a real challenge for researchers. The building, managing, and maintaining complex workflows in a reproducible manner requires extensive knowledge in several areas outside the researchers’ core skill sets, such as software engineering, data science, and high-performance computing (HPC). Our work demonstrates an innovative approach to solving these problems, enabling reproducible EM data analysis through container encapsulated pipelines. Using modern container technologies, we encapsulate processing elements and connect them using shared memory. We expose user-friendly, advanced algorithms and tools to allow end users to utilize without expert programming skills. The platform enables reproducible, scalable, shareable pipelines for the analysis and visualization of EM data. Focusing on interoperability, we leverage the DOE and other agencies’ existing investments to provide a powerful software platform for EM data analysis.

Harris, Christopher↗

DOE OSTI2022

Scalable Gaussian Processes, GPyTorch Application Benchmarking, and Targeted Adaptive Design (TAD) on ThetaGPU

We aim at showcasing the scalability of Gaussian Process (GP). The naive GP implementation scales cubically with data size, which can be prohibitive, so GP has not heretofore been considered suitable for very large-scale problem settings. We take advantage of GPyTorch, a library for scalable GPs built on top of PyTorch that incorporates GPU acceleration. With GPyTorch, one can achieve nearly linear scaling with structured kernel interpolation (SKI) and constant-time predictive covariances computation with LanczOs Variance Estimates (LOVE) while preserving accuracy. We also take advantage of the computational power of ThetaGPU, a supercomputer of Argonne Leadership Computing Facility (ALCF). In addition, we implement a scalable, GPU-ready version of Targeted Adaptive Design (TAD), a GP-based data-driven algorithm that efficiently searches the control space of an advanced manufacturing experiment for settings capable of producing a required design within a specified tolerance, despite the poorly known mapping from control settings to design. We finally show our benchmarking for GPyTorch and TAD performance on CPU vs. ThetaGPU and discuss the results and implications.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS1995

Parallel volume ray-casting for unstructured-grid data on distributed-memory architectures

As computing technology continues to advance, computational modeling of scientific and engineering problems produces data of increasing complexity: large in size and unstructured in shape. Volume visualization of such data is a challenging problem. This paper proposes a distributed parallel solution that makes ray-casting volume rendering of unstructured-grid data practical. Both the data and the rendering process are distributed among processors. At each processor, ray-casting of local data is performed independent of the other processors. The global image composing processes, which require inter-processor communication, are overlapped with the local ray-casting processes to achieve maximum parallel efficiency. This algorithm differs from previous ones in four ways: it is completely distributed, less view-dependent, reasonably scalable, and flexible. Without using dynamic load balancing, test results on the Intel Paragon using from two to 128 processors show, on average, about 60% parallel efficiency.

Ma, Kwan-Liu↗

NASA NTRS2008

Earth Science Mining Web Services

To allow scientists further capabilities in the area of data mining and web services, the Goddard Earth Sciences Data and Information Services Center (GES DISC) and researchers at the University of Alabama in Huntsville (UAH) have developed a system to mine data at the source without the need of network transfers. The system has been constructed by linking together several pre-existing technologies: the Simple Scalable Script-based Science Processor for Measurements (S4PM), a processing engine at he GES DISC; the Algorithm Development and Mining (ADaM) system, a data mining toolkit from UAH that can be configured in a variety of ways to create customized mining processes; ActiveBPEL, a workflow execution engine based on BPEL (Business Process Execution Language); XBaya, a graphical workflow composer; and the EOS Clearinghouse (ECHO). XBaya is used to construct an analysis workflow at UAH using ADam components, which are also installed remotely at the GES DISC, wrapped as Web Services. The S4PM processing engine searches ECHO for data using space-time criteria, staging them to cache, allowing the ActiveBPEL engine to remotely orchestras the processing workflow within S4PM. As mining is completed, the output is placed in an FTP holding area for the end user. The goals are to give users control over the data they want to process, while mining data at the data source using the server's resources rather than transferring the full volume over the internet. These diverse technologies have been infused into a functioning, distributed system with only minor changes to the underlying technologies. The key to the infusion is the loosely coupled, Web-Services based architecture: All of the participating components are accessible (one way or another) through (Simple Object Access Protocol) SOAP-based Web Services.

Pham, Long↗