SEARCH · Engineering Papers
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.
An onboard array structure has been devised for end to end processing of data from multiple spaceborne sensors. The array constitutes sets of programmable pipeline processors whose elements perform each assigned function in 0.25 microseconds. This space shuttle computer system can handle data rates from a few bits to over 100 megabits per second.
To compose the complicated systems using algorithmically specialized logic circuits or processors, one solution is to perform relational computations such as union, division and intersection directly on hardware. These relations can be pipelined efficiently on a network of processors having an array configuration. These processors can be designed and implemented with a few simple cells. In order to determine the state-of-the-art in Electrically Reconfigurable Logic Array (ERLA), a survey of the available programmable logic array (PLA) and the logic circuit elements used in such arrays was conducted. Based on this survey some recommendations are made for ERLA devices.
Scenes containing many polygons generated in real time. Video pipeline subsystem having branched structure performs scan conversion of polygons in images generated by computers. New subsystem divides polygons into triangles, each of which processed rapidly in parallel, modular fashion and merged into image.
The Department of Energy's (DOE) Geothermal Data Repository (GDR) is celebrating its tenth anniversary! Over the last decade it has grown from the simple idea of storing public data in a centralized location to a valuable tool at the center of the US geothermal scientific community and an integral part of the DOE Geothermal Technologies Office (DOE GTO) project management strategy. Researchers funded by the DOE GTO have contributed over 1,300 data submissions to the GDR. These data have been used to further advancements in geothermal science, economic analysis, exploration, research, development, and operational efficiency. The adoption of open data methodologies and a data management strategy that prioritizes universal open access and standardized, interoperable data have further increased the value of GDR data, making them available across a distributed network of data sharing partners and improving their utility to other industries and related fields, including material science and space exploration. Incorporating feedback from users has been critical to the GDR's success, allowing it to grow over the years to meet the evolving needs of the geothermal community. This paper will explore some of many changes that occurred throughout the GDRs tenure and the lessons learned along the way, as well as highlight some of the new features and recent improvements that been implemented to support innovation, reduce duplication of effort, and advance the geothermal industry as a whole.
A digital twin system has been established to model the long term degradation of the state of health of lithium ion batteries. Two data streams result from the computational model of the system and from the physical experiment measurements. A machine learning pipeline has been developed at the nexus of these data streams. Leveraging the unique data sources in multiple transfer learning approaches has lead to the development of multiple cell specific machine learning digital twins. Our discussion on this first of its kind technology will cover challenges in data ingestion, scalability, architecture and orchestration, and research findings.
Accurate metagenomic classification relies on comprehensive, up-to-date, and validated reference databases. While the NCBI BLAST Nucleotide (nt) database, encompassing a vast collection of sequences from all domains of life, represents an invaluable resource, its massive size—currently exceeding 10 12 nucleotides—and exponential growth pose significant challenges for researchers seeking to maintain current nt-based indices for metagenomic classification. Recognizing that no current nt-based indices exist for the widely used Centrifuge classifier, and the last public version currently available was released in 2018, we addressed this critical gap by leveraging advanced high-performance computing resources. We present new Centrifuge-compatible nt databases, meticulously constructed using a novel pipeline incorporating different quality control measures, including reference decontamination and filtering. These measures demonstrably reduce spurious classifications, as shown through our reanalysis of published metagenomic data where Plasmodium annotations were dramatically reduced using our decontaminated database, highlighting how database quality can significantly impact research conclusions. Through temporal comparisons, we also reveal how our approach minimizes inconsistencies in taxonomic assignments stemming from asynchronous updates between public sequence and taxonomy databases. These discrepancies are particularly evident in taxa such as Listeria monocytogenes and Naegleria fowleri, where classification accuracy varied significantly across database versions. These new databases, made available as pre-built Centrifuge indexes, respond to the need for an open, robust, nt-based pipeline for taxonomic classification in metagenomics. Applications such as environmental metagenomics, forensics, and clinical metagenomics, which require comprehensive taxonomic coverage, will benefit from this resource. Our work highlights the importance of treating reference databases as dynamic entities, subject to ongoing quality control and validation akin to software development best practices. This approach is crucial for ensuring accuracy and reliability of metagenomic analysis, especially as databases continue to expand in size and complexity.
The large-scale analysis task of deciphering gravitational wave signals in the LISA data stream will be difficult, requiring a large amount of computational resources and extensive development of computational methods. Its high dimensionality, multiple model types, and complicated noise profile require a global fit to all parameters and input models simultaneously. In this work, we detail our global fit algorithm, called “Erebor,” designed to accomplish this challenging task. It is capable of analysing current state-of-the-art datasets and then growing into the future as more pieces of the pipeline are completed and added. We describe our pipeline strategy, the algorithmic setup, and the results from our analysis of the LDC2A Sangria dataset, which contains Massive Black Hole Binaries, compact Galactic Binaries, and a parameterized noise spectrum whose parameters are unknown to the user. The Erebor algorithm includes three unique and very useful contributions: GPU acceleration for enhanced computational efficiency; ensemble MCMC sampling with multiple MCMC walkers per temperature for better mixing and parallelized sample creation; and special online updates to reversible-jump (or trans-dimensional) sampling distributions to ensure sampler mixing and accurate initial estimates for detectable sources in the data. We recover posterior distributions for all 15 (6) of the injected MBHBs in the LDC2A training (hidden) dataset. We catalog ∼12000 Galactic Binaries (∼8000 as high confidence detections) for both the training and hidden datasets. All of the sources and their posterior distributions are provided in publicly available catalogs.
We report a resource estimation pipeline that explicitly compiles quantum circuits expressed using the Clifford+T gate set into a surface code lattice surgery instruction set. The cadence of magic state requests from the compiled circuit enables the optimization of magic state distillation and storage requirements in a post-hoc analysis. To compile logical circuits into lattice surgery operations, we build upon the open-source Lattice Surgery Compiler. The revised compiler operates in two stages: the first translates logical gates into an abstract, layout-independent instruction set; the second compiles these into local lattice surgery instructions that are allocated to hardware tiles according to a specified resource layout. The second stage retains logical parallelism while avoiding resource contention in the fault-tolerant layer, aiding realism. Additionally, users can specify dedicated tiles at which magic states are replenished, enabling resource costs from the logical computation to be considered independently from magic state distillation and storage. We demonstrate the applicability of our pipeline to large practical quantum circuits by providing resource estimates for the ground state estimation of molecules. Finally, we find that variable magic state consumption rates in real circuits can cause the resource costs of magic state storage to dominate unless production is varied to suit.
Large language models (LLMs) have demonstrated remarkable success as foundational models, benefiting various downstream applications through fine-tuning. Loss scaling studies have demonstrated the superior performance of larger LLMs compared to their smaller counterparts. Nevertheless, training LLMs with billions of parameters poses significant challenges and requires considerable computational resources. For example, training a one trillion parameter GPT-style model on 20 trillion tokens requires a staggering 120 million exaflops. This research explores efficient distributed training strategies to extract this computation from Frontier, the world's first exascale supercomputer. We enable and investigate various model and data parallel training techniques, such as tensor parallelism, pipeline parallelism, and sharded data parallelism, to facilitate training a trillion-parameter model on Frontier. We empirically assess these techniques and their associated parameters to determine their impact on memory footprint, communication latency, and GPU's computational efficiency. We analyze the complex interplay among these techniques and find a strategy to combine them to achieve high throughput through hyperparameter tuning. We have identified efficient strategies for training large LLMs of varying sizes through empirical analysis and hyperparameter tuning. For 22 Billion, 175 Billion, and 1 Trillion parameters, we achieved GPU throughputs of 38.38%, 36.14%, and 31.96%, respectively. For the training of the 175 Billion parameter model and the 1 Trillion parameter model, we achieved 100% weak scaling efficiency on 1024 and 3072 Mi250X GPUs, respectively. We also achieved strong scaling efficiencies of 89% and 87% for these two models. We trained these models only tens of iterations instead of training till completion.
The large-scale analysis task of deciphering gravitational-wave signals in the LISA data stream will be difficult, requiring a large amount of computational resources and extensive development of computational methods. Its high dimensionality, multiple model types, and complicated noise profile require a global fit to all parameters and input models simultaneously. In this work, we detail our global fit algorithm, called “Erebor,” designed to accomplish this challenging task. It is capable of analyzing current state-of-the-art datasets and then growing into the future as more pieces of the pipeline are completed and added. We describe our pipeline strategy, the algorithmic setup, and the results from our analysis of the LDC2A Sangria dataset, which contains massive black hole binaries, compact galactic binaries, and a parametrized noise spectrum whose parameters are unknown to the user. The Erebor algorithm includes three unique and very useful contributions: GPU acceleration for enhanced computational efficiency; ensemble Markov Chain Monte Carlo (MCMC) sampling with multiple MCMC walkers per temperature for better mixing and parallelized sample creation; and special online updates to reversible-jump (or transdimensional) sampling distributions to ensure sampler mixing and accurate initial estimates for detectable sources in the data.We recover posterior distributions for all 15 (6) of the injected massive black hole binaries (MBHB) in the LDC2A training (hidden) dataset. We catalog ∼12000 galactic binaries (∼8000 as high confidence detections) for both the training and hidden datasets. All of the sources and their posterior distributions are provided in publicly available catalogs.
Here, the effectiveness of continuous vibro-impact forcing representations for the cantilevered pipe that conveys fluid is explored and analyzed. The previously accepted forcing model utilizing a smoothened trilinear spring is estimated using three continuous forcing representations, namely, polynomial, rational polynomial, and hyperbolic tangent. The accuracy of the estimated forcing functions is investigated and analyzed by calculating the root mean square error, and bifurcation diagrams are generated and compared to the nominal system. Additionally, the dynamic response of the system is further characterized using Poincare maps, power spectra, and basins of attraction. Once all continuous forcing representations are analyzed and compared to the nominal system, the computational cost of each method is examined, and further limitations of the hyperbolic tangent method are discovered. It is proved that the hyperbolic tangent forcing representation most accurately captures the dynamic response of the pipeline, and the least accurate representation is the rational polynomial representation. Additionally, considerable computational cost is saved when employing the hyperbolic tangent representation compared to the discontinuous representation.
As global interest in renewable energy continues to increase, there has been a pressing need for developing novel energy storage devices based on organic electrode materials that can overcome the shortcomings of the current lithium-ion batteries. One critical challenge for this quest is to find materials whose redox potential (RP) meets specific design targets. In this study, we propose a computational framework for addressing this challenge through the effective design and optimal operation of a high-throughput virtual screening (HTVS) pipeline that enables rapid screening of organic materials that satisfy the desired criteria. Starting from a high-fidelity model for estimating the RP of a given material, we show how a set of surrogate models with different accuracy and complexity may be designed to construct a highly accurate and efficient HTVS pipeline. We demonstrate that the proposed HTVS pipeline construction and operation strategies substantially enhance the overall screening throughput.
Techniques for partitioning optical linear algebra problems to make them amenable to solution using optical processors programmed with simple algorithms are explored. Generalized methods for splitting a linear algebra matrix into a series of submatrices are reviewed, showing that simple forms can be pipelined smoothly and that parallel accumulation can be achieved by beam combining on detectors or by summing electronically. The techniques offer simplified bookkeeping, algorithmic independence, and high efficiency. The computational speed will depend on the number of multiplier-accumulators devoted to the task.
X-ray absorption spectroscopy (XAS) is a powerful characterization technique for probing the local chemical environment of absorbing atoms. However, analyzing XAS data presents significant challenges, often requiring extensive, computationally intensive simulations, as well as significant domain expertise. These limitations hinder the development of fast, robust XAS analysis pipelines that are essential in high-throughput studies and for autonomous experimentation. Here, we address these challenges with OmniXAS, a framework that contains a suite of transfer learning approaches for XAS prediction, each uniquely contributing to improved accuracy and efficiency, as demonstrated on the K-edge spectra database covering eight 3d transition metals (Ti–Cu). The OmniXAS framework is built upon three distinct strategies. First, we use M3GNet [Nat. Comput. Sci. 2, 718 (2022)] to derive latent representations of the local chemical environment of absorption sites as input for XAS prediction, achieving significant improvements over conventional featurization techniques. Second, we employ a hierarchical transfer learning strategy, training a universal multitask model across elements before fine-tuning for element-specific predictions. Models based on this cascaded approach after elementwise fine-tuning outperform element-specific models by up to 69%. Third, we implement cross-fidelity transfer learning, adapting a universal model to predict spectra generated by simulation of a different fidelity with a much higher computational cost. This approach improves prediction accuracy by up to 11% over models trained on the target fidelity alone. Our approach significantly boosts the throughput of XAS modeling by orders of magnitude as compared to first-principles simulations and is extendable to XAS prediction for a broader range of elements. The proposed transfer learning framework is generalizable to enhance deep-learning models that target other properties in materials research.
We present our search for strong lens, galaxy-scale systems in the first data release of the Dark Energy Survey (DES), based on a color-selected parent sample of 18 745 029 luminous red galaxies (LRGs). We used a convolutional neural network (CNN) to grade this LRG sample with values between 0 (non-lens) and 1 (lens). Our training set of mock lenses is data-driven, that is, it uses lensed sources taken from HST-COSMOS images and lensing galaxies from DES images of our LRG sample. A total of 76 582 cutouts were obtained with a score above 0.9, which were then visually inspected and classified into two catalogs. The first one contains 405 lens candidates, of which 90 present clear lensing features and counterparts, while the other 315 require more evidence, such as higher resolution imaging or spectra, to be conclusive. A total of 186 candidates are newly identified by our search, of which 16 are among the 90 most promising (best) candidates. The second catalog includes 539 ring galaxy candidates. This catalog will be a useful false positive sample for training future CNNs. For the 90 best lens candidates we carry out color-based deblending of the lens and source light without fitting any analytical profile to the data. This method is shown to be very efficient in the deblending, even for very compact objects and for objects with a complex morphology. Finally, from the 90 best lens candidates, we selected 52 systems with one single deflector to test an automated modeling pipeline that has the capacity to successfully model 79% of the sample within an acceptable computing runtime.
The VA Environmental Determinants of Health (EDH) Advanced Software Pipeline Framework is designed to enhance the efficiency, scalability, and security of geospatial data processing workflows. This framework integrates modern data orchestration and containerization technologies, including Prefect for workflow automation, Docker for containerization, and PostgreSQL/PostGIS for geospatial data storage and analysis. It ensures standardized, reproducible, and automated data processing, supporting VA objectives related to substance use risk assessment and recovery research. The pipeline addresses key scalability and performance challenges through horizontal and vertical scaling, high-performance computing (HPC) integration, parallel processing, task caching, and dynamic resource allocation. These optimizations improve throughput and reduce latency, allowing the system to efficiently manage large and complex datasets. Additionally, security and compliance measures—such as data encryption (SSL), Role-Based Access Control (RBAC), and adherence to GDPR and HIPAA standards—safeguard sensitive information throughout data transmission and storage. A key implementation of this framework includes the automation of shelter list geolocation workflows, ensuring that up-to-date data is readily available for VA decision-making. Lessons learned from this project include the transition from in-memory processing to incremental storage writes, improving resource management and reliability. Future enhancements aim to expand automation, integrate AI-driven anomaly detection, and incorporate high-performance computing resources. This framework provides a scalable, secure, and adaptable solution for managing geospatial datasets, reinforcing the VA’s ability to support clinical and strategic initiatives through data-driven decision-making.
Computational algorithms are presented for the finite element dynamic analysis of structures on the CDC STAR-100 computer. The spatial behavior is described using higher-order finite elements. The temporal behavior is approximated by using either the central difference explicit scheme or Newmark's implicit scheme. In each case the analysis is broken up into a number of basic macro-operations. Discussion is focused on the organization of the computation and the mode of storage of different arrays to take advantage of the STAR pipeline capability. The potential of the proposed algorithms is discussed and CPU times are given for performing the different macro-operations for a shell modeled by higher order composite shallow shell elements having 80 degrees of freedom.
The philosophy and data encoding possible in systolic array optical processor (SAOP) were reviewed. The multitude of linear algebraic operations achievable on this architecture is examined. These operations include such linear algebraic algorithms as: matrix-decomposition, direct and indirect solutions, implicit and explicit methods for partial differential equations, eigenvalue and eigenvector calculations, and singular value decomposition. This architecture can be utilized to realize general techniques for solving matrix linear and nonlinear algebraic equations, least mean square error solutions, FIR filters, and nested-loop algorithms for control engineering applications. The data flow and pipelining of operations, design of parallel algorithms and flexible architectures, application of these architectures to computationally intensive physical problems, error source modeling of optical processors, and matching of the computational needs of practical engineering problems to the capabilities of optical processors are emphasized.