Search NASA⌕ Search

SEARCH · Search NASA

Results for “program architectures”

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 1,009 records · Page 56

Scalability Tests of a Finite Element Code on Hundreds of Thousands Cores and Heterogeneous Architecture

A multi-scale finite element method code, msFEM, is tested on Jaguar and Nebulae, two petaflops computers that were listed as #1 and #2 on the Top500 list of June 2010 at the time of the tests. The flat MPI version of msFEM is scaled from 20K up to 200K CPU cores on Jaguar, delivering 70% parallel efficiency at the 200K cores with a finite element model of eight millions of degrees of freedom. GPU versions, in both double precision and mixed precision coded through MPI+OpenMP+CUDA hybrid programming, 900 GPU nodes on Jaguar and 1500 GPU nodes on Nebulae, achieving remarkable 90 + % parallel efficiency on the systems. The mixed-precision GPU version delivers further 1.5 times of speedup over the fully double precision version with no significant implementational cost. The large-scale tests support that the msFEM runs efficiently on petaflops computers and is highly potential for domain applications at extreme-scale.

finite element method↗

Nuclear, Land marker for a Waste Isolation Site: International Architecture Competition - 18240

The design and construction of a nuclear waste marker must include many architectural considerations. Very little research has been conducted by architects to address this issue to date. Historically, the task of designating a nuclear waste site has been a discussion predominantly contained within the scientific and engineering communities. In an effort to engage architects within this dialogue, arch out loud hosted the NUCLEAR competition as an international, architectural investigation into the design of a nuclear waste site marker. The NUCLEAR competition asks participants to design a marker to designate the site of the Waste Isolation Pilot Plant (WIPP) located near Carlsbad, New Mexico. The marker will be meant to be a means of passive institutional controls (PICs) to deter inadvertent human intrusion upon the site and should last 10,000 years. The NUCLEAR competition defined the purpose of the nuclear waste site marker as a means to communicate the danger of inadvertent intrusion from future generations. The competition requires that entries must be both legible in communication and durable in their construction. The competition brief challenged participants to consider how customs and languages will evolve over time and to predict what motivations future generations might have for digging at the site. Other background information was provided about the site such as how waste was stored and how the unique geology of the site seals the waste. The 1993 Sandia National Laboratories report titled Expert Judgment on Markers to Deter Inadvertent Human Intrusion into the Waste Isolation Pilot Plant [1] was referenced along with the 1992 WIPP Land Withdrawal Act [2], interviews with WIPP scientists [3] involved with the WIPP PICs program and the Nuclear Energy Agency Records Knowledge and Memory working group. Competition participants were given digital 3-D Computer-Aided Design models for reference that included the underground storage portion of the site, existing buildings and the 16-square-mile site boundary identified in the 1992 WIPP Land Withdrawal Act [2]. This paper will: - Provide a detailed description of the submitted proposals including graphic analysis of the visualizations, diagrams, technical drawings, jury comments and statements. - Catalogue and identify the consistent themes within all submissions with possible categories including: 1) overall architectural form and language, 2) integration of written language and symbols within architecture, 3) use of technology and robotics, 4) construction methods and materials. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Solid Oxide Fuel Cell Systems PVL Line

In July 2010, Stark State College (SSC), received Grant DE-EE0003229 from the U.S. Department of Energy (DOE), Golden Field Office, for the development of the electrical and control systems, and mechanical commissioning of a unique 20kW scale high-pressure, high temperature, natural gas fueled Stack Block Test System (SBTS). SSC worked closely with subcontractor, Rolls-Royce Fuel Cell Systems (US) Inc. (RRFCS) over a 13 month period to successfully complete the project activities. This system will be utilized by RRFCS for pre-commercial technology development and training of SSC student interns. In the longer term, when RRFCS is producing commercial products, SSC will utilize the equipment for workforce training. In addition to DOE Hydrogen, Fuel Cells, and Infrastructure Technologies program funding, RRFCS internal funds, funds from the state of Ohio, and funding from the DOE Solid State Energy Conversion Alliance (SECA) program have been utilized to design, develop and commission this equipment. Construction of the SBTS (mechanical components) was performed under a Grant from the State of Ohio through Ohio's Third Frontier program (Grant TECH 08-053). This Ohio program supported development of a system that uses natural gas as a fuel. Funding was provided under the Department of Energy (DOE) Solid-state Energy Conversion Alliance (SECA) program for modifications required to test on coal synthesis gas. The subject DOE program provided funding for the electrical build, control system development and mechanical commissioning. Performance testing, which includes electrical commissioning, was subsequently performed under the DOE SECA program. Rolls-Royce Fuel Cell Systems is developing a megawatt-scale solid oxide fuel cell (SOFC) stationary power generation system. This system, based on RRFCS proprietary technology, is fueled with natural gas, and operates at elevated pressure. A critical success factor for development of the full scale system is the capability to test fuel cell components at a scale and under conditions that can be accurately extrapolated to full system performance. This requires specially designed equipment that replicates the pressure (up to 6.5 bara), temperature (about 910 C), anode and cathode gas compositions, flows and power generation density of the full scale design. The SBTS fuel cell anode gas is produced through the reaction of pipeline natural gas with a mixture of steam, CO2, and O2 in a catalytic partial oxidation (CPOX) reactor. Production of the fuel cell anode gas in this manner provides the capability to test a fuel cell with varying anode gas compositions ranging from traditional reformed natural gas to a coal-syngas surrogate fuel. Stark State College and RRFCS have a history of collaboration. This is based upon SSCAs commitment to provide students with skills for advanced energy industries, and RRFCS need for a workforce that is skilled in high temperature fuel cell development and testing. A key to this approach is the access of students to unique SOFC test and evaluation equipment. This equipment is designed and developed by RRFCS, with the participation of SSC interns. In the near-term, the equipment will be used by RRFCS for technology development. When this stage is completed, and RRFCS has moved to commercial products, SSC will utilize this equipment for workforce training. The RRFCS fuel cell design is based upon a unique ceramic substrate architecture in which a porous, flat substrate (tube) provides the support structure for a network of solid oxide fuel cells that are electrically connected in series. These tubes are grouped into a {approx}350-tube repeat configuration, called a stack/block. Stack/block testing, performed at system conditions, provides data that can be confidently scaled to full scale performance. This is the basis for the specially designed and developed test equipment that is required for advancing and accelerating the RRFCS SOFC power system development program. All contract DE-EE0003229 objectives were achieved and deliverables completed during the period from March 1, 2010 through March 31, 2011. As a result of program completion, the Stack Block Test System was ready to support installation and electrical operation of a RRFCS solid oxide fuel cell (SOFC) stack block in the second quarter of 2011.

01 COAL, LIGNITE, AND PEAT↗

Robustness of Deep Learning Classification to Adversarial Input on GPUs: Asynchronous Parallel Accumulation Is a Source of Vulnerability

The ability of machine learning (ML) classification models to resist small, targeted input perturbations—known as adversarial attacks—is a key measure of their safety and reliability. We show that floating-point non associativity (FPNA) coupled with asynchronous parallel programming on GPUs is sufficient to result in misclassification, without any perturbation to the input. Additionally, we show that this misclassification is particularly significant for inputs close to the decision boundary and that standard adversarial robustness results may be overestimated up to 4.6 when not considering machine-level details. We first study a linear classifier, before focusing on standard Graph Neural Network (GNN) architectures and datasets used in robustness assessments. We develop a novel black-box attack using Bayesian optimization to discover external workloads that can change the instruction scheduling which bias the output of reductions on GPUs and reliably lead to misclassification. Motivated by these results, we present a new learnable permutation (LP) gradient-based approach to learning floating-point operation orderings that lead to misclassifications. The LP approach provides a worst-case estimate in a computationally efficient manner, avoiding the need to run identical experiments tens of thousands of times over a potentially large set of possible GPU states or architectures. Finally, using instrumentation-based testing, we investigate parallel reduction ordering across different GPU architectures under external background workloads, when utilizing multi-GPU virtualization, and when applying power capping. Our results demonstrate that parallel reduction ordering varies significantly across architectures under the first two conditions, substantially increasing the search space required to fully test the effects of this parallel scheduler-based vulnerability. These results and the methods developed here can help to include machine-level considerations into adversarial robustness assessments, which can make a difference in safety and mission critical applications.

Shanmugavelu, Sanjif [Maxeler Technologies, a Groq↗

Experiences with user-centered design for the tigres workflow API

Scientific data volumes have been growing exponentially. This has resulted in the need for new tools that enable users to operate on and analyze data. Cyber infrastructure tools, including workflow tools, that have been developed in the last few years has often fallen short if user needs and suffered from lack of wider adoption. User-centered Design (UCD) process has been used as an effective approach to develop usable software with high adoption rates. However, UCD has largely been applied for user-interfaces and there has been limited work in applying UCD to application program interfaces and cyber infrastructure tools. We use an adapted version of UCD that we refer to as Scientist-Centered Design (SCD) to engage with users in the design and development of Tigres, a workflow application programming interface. Tigres provides a simple set of programming templates (e.g., sequence, parallel, split, merge) that can be can used to compose and execute computational and data transformation pipelines. In this paper, we describe Tigres and discuss our experiences with the use of UCD for the initial development of Tigres. Our experience-to-date is that the UCD process not only resulted in better requirements gathering but also heavily influenced the architecture design and implementation details. User engagement during the development of tools such as Tigres is critical to ensure usability and increase adoption.

Ramakrishnan, L↗

Domain-decomposition nonlinear manifold reduced order model

This software combines nonlinear-manifold reduced order models (NM-ROMs) with domain decomposition (DD) techniques. NM-ROMs, which utilize a shallow, sparse autoencoder trained with full order model (FOM) snapshot data, approximate the FOM state on a nonlinear manifold. These models offer advantages over linear-subspace ROMs (LS-ROMs) particularly in scenarios with slowly decaying Kolmogorov n-width. However, the training of NM-ROMs involves a number of parameters that scale with the size of the FOM, and storing high-dimensional FOM snapshots can significantly increase the cost of ROM training for extreme-scale problems. To mitigate these costs, the software employs DD to partition the FOM into smaller subdomains, computes NM-ROMs for each, and then integrates these to form a global NM-ROM. This strategy offers multiple benefits: it enables parallel training of subdomain NM-ROMs, reduces the number of parameters needed, decreases the dimensional requirements of subdomain FOM training data, and allows for customization to the unique characteristics of each FOM subdomain. The use of a shallow, sparse autoencoder architecture in each subdomain NM-ROM facilitates the application of hyper-reduction (HR), simplifying the nonlinear complexities and enhancing computational speed. This software marks the inaugural application of NM-ROM combined with HR to a DD problem. It features an algebraic DD reformulation of the FOM, training of NM-ROMs with HR for each subdomain, and employs a sequential quadratic programming (SQP) solver for the evaluation of the coupled global NMROM. The effectiveness of the DD NM-ROM with HR is numerically demonstrated on the 2D steady-state Burgers' equation, showing an order of magnitude improvement in accuracy over the DD LS-ROM with HR.

Diaz, AlejandroN↗

Energy Efficiency Improvements to Wundar Hall, a Historic Building on the Concordia Campus, Milwaukee, Wisconsin

The Forest County Potawatomi Community (FCPC or Community) implemented energy efficiency improvements to revitalize Wundar Hall, a 34,000 square foot (SF) building that was formerly used as a dormitory and is listed on the National Registry of Historic Places, into an office building. Wundar Hall is the first of many architecturally and historically significant buildings that the Community hopes to renovate at the former Concordia College campus, property on the near west side of Milwaukee that was taken into trust for the Community by the United States on July 10, 1990 (collectively, the Concordia Trust Property). As part of this project, which was conducted with assistance from the Department of Energy's Tribal Energy Program (TEP), the Community updated and/or replaced the building envelope, mechanical systems, the plumbing system, the electrical infrastructure, and building control systems. The project is expected to reduce the building's natural gas consumption by 58% and the electricity consumption by 55%. In addition, the project was designed to act as a catalyst to further renovation of the Concordia Trust Property and the neighborhood. The City of Milwaukee has identified redevelopment of the Concordia Trust Property as a Catalytic Project for revitalizing the near west side. The Tribe envisions a revitalized, mixed-use campus of community services, education, and economic development—providing services to the Indian community and jobs to the neighborhood.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

AI Benchmark Democratization and Carpentry

Benchmarks are a cornerstone of modern machine learning, enabling reproducibility, comparison, and scientific progress. However, AI benchmarks are increasingly complex, requiring dynamic, AI-focused workflows. Rapid evolution in model architectures, scale, datasets, and deployment contexts makes evaluation a moving target. Large language models often memorize static benchmarks, causing a gap between benchmark results and real-world performance. Beyond traditional static benchmarks, continuous adaptive benchmarking frameworks are needed to align scientific assessment with deployment risks. This calls for skills and education in AI Benchmark Carpentry. From our experience with MLCommons, educational initiatives, and programs like the DOE's Trillion Parameter Consortium, key barriers include high resource demands, limited access to specialized hardware, lack of benchmark design expertise, and uncertainty in relating results to application domains. Current benchmarks often emphasize peak performance on top-tier hardware, offering limited guidance for diverse, real-world scenarios. Benchmarking must become dynamic, incorporating evolving models, updated data, and heterogeneous platforms while maintaining transparency, reproducibility, and interpretability. Democratization requires both technical innovation and systematic education across levels, building sustained expertise in benchmark design and use. Benchmarks should support application-relevant comparisons, enabling informed, context-sensitive decisions. Dynamic, inclusive benchmarking will ensure evaluation keeps pace with AI evolution and supports responsible, reproducible, and accessible AI deployment. Community efforts can provide a foundation for AI Benchmark Carpentry.

von Laszewski, Gregor [Virginia U.]↗

mpi-sppy: Optimization Under Uncertainty for Pyomo

We introduce a new Pyomo extension for optimization under uncertainty, mpi-sppy. This extension allows for the mixing of, and sharing of information between, various algorithmic approaches for optimization under uncertainty. We will discuss the underlying architecture and assess the performance and scalability of mpi-sppy on various systems, including HPC clusters.

27 ARPA - Advanced Research Projects Agency-Energy↗

Numerical simulation: Toward the design of high-efficiency planar perovskite solar cells

Organo-metal halide perovskite solar cells based on planar architecture have been reported to achieve remarkably high power conversion efficiency (PCE, >16%), rendering them highly competitive to the conventional silicon based solar cells. A thorough understanding of the role of each component in solar cells and their effects as a whole is still required for further improvement in PCE. In this work, the planar heterojunction-based perovskite solar cells were simulated with the program AMPS (analysis of microelectronic and photonic structures)-1D. Simulation results revealed a great dependence of PCE on the thickness and defect density of the perovskite layer. Meanwhile, parameters including the work function of the back contact as well as the hole mobility and acceptor density in hole transport materials were identified to significantly influence the performance of the device. Strikingly, an efficiency over 20% was obtained under the moderate simulation conditions.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Tiling Framework for Heterogeneous Computing of Matrix based Tiled Algorithms

Tiling matrix operations can improve the load balancing and performance of applications on heterogeneous computing resources. Writing a tile-based algorithm for each operation with a traditional, hand-tuned tiling approach that uses for loops in C/C++ is cumbersome and error prone. Moreover, it must enable and support the heterogeneous memory management of data objects and also explore architecture-supported, native, tiled-data transfer APIs instead of copying the tiled data to continuous memory before the data transfer. The tiling framework provides a tiled data structure for heterogeneous memory mapping and parameterization to a heterogeneous task specification API. We have integrated our tiled framework into MatRIS (Math kernels library using IRIS). IRIS is a heterogeneous run-time framework with a heterogeneous programming model, memory model, and task execution model. Experiments reveal that the tiled framework for BLAS operations has improved the programmability of tiled BLAS and improved performance by ~20% when compared against the traditional method that copies the data to continuous memory locations for heterogeneous computing.

Miniskar, Narasinga Rao↗

Developing an Ultra-Compact, Topology-Optimized Heat Exchanger using Additive Manufacturing

Traditional heat exchanger designs have relied on a plate-fin architecture, assembled in a crossflow configuration. These HX designs are limited in their maximum allowable temperature and, therefore, efficiency. More advanced designs, such as those containing counter-flow, microtube cores, can yield higher efficiencies, but generally require complex headers. The design of such headers is usually determined based on user experience rather than on rigorous design tools, which can lead to lossy architectures; conventional headers can generate up to 50% of the total pressure loss through the HX. Furthermore, the consequences of manufacturing techniques are rarely considered during the design phase. Our team has developed an ultra-compact, topology-optimized heat exchanger capable of operating in environments with temperatures and pressures up to 800°C (1472°F) and 250 bar (3626 psi), respectively. Throughout the program, the team leveraged extensive additive manufacturing research and experience in supercritical carbon dioxide (sCO 2 ) power generation experience to develop and commercialize the technology. The final power density numbers for the subscale and full-scale heat exchangers are >15 kW/kg and >50 kW/L, which includes the core and the headers. The team has involved the RTX business units, notably Collins Aerospace, in discussions for commercializing the technology.

20 FOSSIL-FUELED POWER PLANTS↗

GPU-accelerated 3D neutron diffusion code based on finite difference method

Finite difference method, as a traditional numerical solution to neutron diffusion equation, although considered simpler and more precise than the coarse mesh nodal methods, has a bottle neck to be widely applied caused by the huge memory and unendurable computation time it requires. In recent years, the concept of General-Purpose computation on GPUs has provided us with a powerful computational engine for scientific research. In this study, a GPU-Accelerated multi-group 3D neutron diffusion code based on finite difference method was developed. First, a clean-sheet neutron diffusion code (3DFD-CPU) was written in C++ on the CPU architecture, and later ported to GPUs under NVIDIA's CUDA platform (3DFD-GPU). The IAEA 3D PWR benchmark problem was calculated in the numerical test, where three different codes, including the original CPU-based sequential code, the HYPRE (High Performance Pre-conditioners)-based diffusion code and CITATION, were used as counterpoints to test the efficiency and accuracy of the GPU-based program. The results demonstrate both high efficiency and adequate accuracy of the GPU implementation for neutron diffusion equation. A speedup factor of about 46 times was obtained, using NVIDIA's Geforce GTX470 GPU card against a 2.50 GHz Intel Quad Q9300 CPU processor. Compared with the HYPRE-based code performing in parallel on an 8-core tower server, the speedup of about 2 still could be observed. More encouragingly, without any mathematical acceleration technology, the GPU implementation ran about 5 times faster than CITATION which was speeded up by using the SOR method and Chebyshev extrapolation technique. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Architecture-Level Dependability Analysis of a Medical Decision Support System

Recent advances in techniques such as image analysis, text analysis and machine learning have shown great potential to assist physicians in detecting and diagnosing health issues in patients. In this paper, we describe the approach and findings of an architecture-level dependability analysis for a mammography decision support system that incorporates these techniques. The goal of the research described in this paper is to provide an initial understanding of the dependability issues, particularly the potential failure modes and severity, in order to identify areas of potential high risk. The results will guide design decisions and provide the basis of a dependability and performance evaluation program.

99 GENERAL AND MISCELLANEOUS↗

Notified access in coarray-based hydrodynamics applications on many-core architectures: Design and performance

With the increasing availability of the Remote Direct Memory Access (RDMA) support in computer networks, the so called Partitioned Global Address Space (PGAS) model has evolved in the last few years. Although there are several cases where a PGAS approach can easily solve difficult message passing situations, like in particle tracking and adaptive mesh refinement applications, the producer-consumer pattern, usually adopted in task-based parallelism, can only be implemented inefficiently because of the separation between data transfer and synchronization (which is usually unified in message passing programming models). Here, in this paper, we provide two contributions: (1) we propose an extension for the Fortran language that provides the concept of Notified Access by associating regular coarray variables with event variables. (2) We demonstrate that the MPI extension proposed by foMPI for Notified Access can be used effectively to implement the same concept in a PGAS run-time library like OpenCoarrays. Moreover, for a hydrodynamics mini-application, we found that Fortran 2018 events perform always better than Fortran 2008 sync statements on many-core processors. We finally show how the proposed Notified Access can improve the performance even more.

97 MATHEMATICS AND COMPUTING↗

An interactive machine learning platform for analyzing multi-particle coincidence data from cold target recoil ion momentum spectroscopy

We present SCULPT (Supervised Clustering and Uncovering Latent Patterns with Training), a comprehensive software platform for analyzing tabulated high-dimensional multi-particle coincidence data from Cold Target Recoil Ion Momentum Spectroscopy (COLTRIMS) experiments. The software addresses critical challenges in modern momentum spectroscopy by integrating advanced machine learning techniques with physics-informed analysis in an interactive web-based environment. SCULPT implements uniform manifold approximation and projection for non-linear dimensionality reduction to reveal correlations in high-dimensional data. We also discuss potential extensions to deep autoencoders for feature learning and genetic programming for automated discovery of physically meaningful observables. A novel adaptive confidence scoring system provides quantitative reliability assessments by evaluating user-selected clustering quality metrics with predefined weights that reflect each metric’s robustness. The platform features configurable molecular profiles for different experimental systems, interactive visualization with selection tools, and comprehensive data filtering capabilities. Utilizing a subset of SCULPT’s capabilities, we analyze photo-double-ionization data measured using the COLTRIMS method for three-body dissociation of the D 2 O molecule, revealing distinct fragmentation channels and their correlations with physics parameters. The software’s modular architecture and web-based implementation make it accessible to the broader atomic and molecular physics community, significantly reducing the time required for complex multi-dimensional analyses. This opens the door to finding and isolating rare events exhibiting non-linear correlations on the fly during experimental measurements, which can help steer exploration and improve the efficiency of experiments.

Artificial neural networks↗

Static analysis techniques for semiautomatic synthesis of message passing software skeletons

The design of high-performance computing architectures demands performance analysis of large-scale parallel applications to derive various parameters concerning hardware design and software development. The process of performance analysis and benchmarking an application can be done in several ways with varying degrees of fidelity. One of the most cost-effective ways is to do a coarse-grained study of large-scale parallel applications through the use of program skeletons. The concept of a “program skeleton” that we discuss in this article is an abstracted program that is derived from a larger program where source code that is determined to be irrelevant is removed for the purposes of the skeleton. In this work, we develop a semiautomatic approach for extracting program skeletons based on compiler program analysis. Finally, we demonstrate correctness of our skeleton extraction process by comparing details from communication traces, as well as show the performance speedup of using skeletons by running simulations in the SST/macro simulator.

97 MATHEMATICS AND COMPUTING↗