Search NASA⌕ Search

SEARCH · Search NASA

Results for “Failure Rate”

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 487 records · Page 27

A Model-Based Case for Redundant Computation

Despite its seemingly nonsensical cost, we show through modeling and simulation that redundant computation merits full consideration as a resilience strategy for next-generation systems. Without revolutionary breakthroughs in failure rates, part counts, or stable-storage bandwidths, it has been shown that the utility of Exascale systems will be crushed by the overheads of traditional checkpoint/restart mechanisms. Alternate resilience strategies must be considered, and redundancy is a proven unrivaled approach in many domains. We develop a distribution-independent model for job interrupts on systems of arbitrary redundancy, adapt Daly’s model for total application runtime, and find that his estimate for optimal checkpoint interval remains valid for redundant systems. We then identify conditions where redundancy is more cost effective than non-redundancy. These are done in the context of the number one supercomputers of the last decade, showing that thorough consideration of redundant computation is timely - if not overdue.

97 MATHEMATICS AND COMPUTING↗

FOX: A Fault-Oblivious Extreme-Scale Execution Environment Boston University Final Report Project Number: DE-SC0005365

Exascale computing systems will provide a thousand-fold increase in parallelism and a proportional increase in failure rate relative to today's machines. Systems software for exascale machines must provide the infrastructure to support existing applications while simultaneously enabling efficient execution of new programming models that naturally express dynamic, adaptive, irregular computation; coupled simulations; and massive data analysis in a highly unreliable hardware environment with billions of threads of execution. The FOX project explored systems software and runtime support for a new approach to the data and work distribution for fault oblivious application execution. Our major OS work at Boston University focused on developing a new light-weight operating systems model that provides an appropriate context for both multi-core and multi-node application development. This work is discussed in section 1. Early on in the FOX project BU developed infrastructure for prototyping dynamic HPC environments in which the sets of nodes that an application is run on can be dynamically grown or shrunk. This work was an extension of the Kittyhawk project and is discussed in section 2. Section 3 documents the publications and software repositories that we have produced. To put our work in context of the complete FOX project contribution we include in section 4 an extended version of a paper that documents the complete work of the FOX team.

97 MATHEMATICS AND COMPUTING↗

Report for the ASC CSSE L2 Milestone (4873) - Demonstration of Local Failure Local Recovery Resilient Programming Model.

Recovery from process loss during the execution of a distributed memory parallel application is presently achieved by restarting the program, typically from a checkpoint file. Future computer system trends indicate that the size of data to checkpoint, the lack of improvement in parallel file system performance and the increase in process failure rates will lead to situations where checkpoint restart becomes infeasible. In this report we describe and prototype the use of a new application level resilient computing model that manages persistent storage of local state for each process such that, if a process fails, recovery can be performed locally without requiring access to a global checkpoint file. LFLR provides application developers with an ability to recover locally and continue application execution when a process is lost. This report discusses what features are required from the hardware, OS and runtime layers, and what approaches application developers might use in the design of future codes, including a demonstration of LFLR-enabled MiniFE code from the Matenvo mini-application suite.

Heroux, Michael Allen↗

CHP Fuel Cell Durability Demonstration - Final Report

Plug Power has managed a demonstration project that has tested multiple units of its high-temperature, PEM fuel cell system in micro-combined heat and power (μ-CHP) applications in California. The specific objective of the demonstration project was to substantiate the durability of GenSys Blue, and, thereby, verify its technology and commercial readiness for the marketplace. In the demonstration project, Plug Power, in partnership with the National Fuel Cell Research Center (NFCRC) at the University of California, Irvine (UCI), and Sempra, will execute two major tasks: • Task 1: Internal durability/reliability fleet testing. Six GenSys Blue units will be built and will undergo an internal test regimen to estimate failure rates. This task was modified to include 3 GenSys Blue units installed in a lab at UCI. • Task 2: External customer testing. Combined heat and power units will be installed and tested in real-world residential and/or light commercial end user locations in California.

Petrecky, James↗

FOX: A Fault Oblivious Extreme-Scale Execution Environment

The FOX project, funded under the ASCR X-stack I program, developed systems software and runtime libraries for a new approach to the data and work distribution for massively parallel, fault oblivious application execution. Our work was motivated by the premise that exascale computing systems will provide a thousand-fold increase in parallelism and a proportional increase in failure rate relative to today’s machines. To deliver the capability of exascale hardware, the systems software must provide the infrastructure to support existing applications while simultaneously enabling efficient execution of new programming models that naturally express dynamic, adaptive, irregular computation; coupled simulations; and massive data analysis in a highly unreliable hardware environment with billions of threads of execution. Our OS research has prototyped new methods to provide efficient resource sharing, synchronization, and protection in a many-core compute node. We have experimented with alternative task/dataflow programming models and shown scalability in some cases to hundreds of thousands of cores. Much of our software is in active development through open source projects. Concepts from FOX are being pursued in next generation exascale operating systems. Our OS work focused on adaptive, application tailored OS services optimized for multi → many core processors. We developed a new operating system NIX that supports role-based allocation of cores to processes which was released to open source. We contributed to the IBM FusedOS project, which promoted the concept of latency-optimized and throughput-optimized cores. We built a task queue library based on distributed, fault tolerant key-value store and identified scaling issues. A second fault tolerant task parallel library was developed, based on the Linda tuple space model, that used low level interconnect primitives for optimized communication. We designed fault tolerance mechanisms for task parallel computations employing work stealing for load balancing that scaled to the largest existing supercomputers. Finally, we implemented the Elastic Building Blocks runtime, a library to manage object-oriented distributed software components. To support the research, we won two INCITE awards for time on Intrepid (BG/P) and Mira (BG/Q). Much of our work has had impact in the OS and runtime community through the ASCR Exascale OS/R workshop and report, leading to the research agenda of the Exascale OS/R program. Our project was, however, also affected by attrition of multiple PIs. While the PIs continued to participate and offer guidance as time permitted, losing these key individuals was unfortunate both for the project and for the DOE HPC community.

97 MATHEMATICS AND COMPUTING↗

High Thermal Conductivity, Radiation Resistant Adhesive for High Energy Physics

High Energy Physics systems include particle detector electronics that are integrated with cooling channels and structural housings. Adhesives used to bond these devices to the cooling lines and housings should contribute to cooling by conducting heat away from the electronics and preventing hot spots from developing in either the electronics or the housing. Effective cooling can produce order of magnitude reduction in electronics failure rates. During Phase I, Surmet demonstrated feasibility of a novel radiation-tolerant high-thermal-conductivity adhesive for HEP systems. This was achieved through use of suitable high thermal conductivity fillers to resin adhesive. Feasibility of the technology was demonstrated through key processing and characterization/testing work. The novel filled adhesive was produced in thin-film form to promote heat removal performance and simplify assembly. It was shown to have thermal conductivity >35x that of conventional epoxies. It was shown to successfully withstand strong gamma and proton irradiation. This was demonstrated through density and microstructural characterization as well as thermal conductivity and strength measurement, before and after irradiation. Finally, production of the novel filled adhesive relies on mature processing techniques, simplifying the transition to scale-up and commercialization. During the course of Phase I, Surmet held discussions with a number of researchers working on DOE HEP systems, to understand requirements. These discussions guided the Phase I research, and will be used to plan out follow-on Phase II work. In Phase II research, we will optimize the filled adhesive and tailor it to various product forms to address the multiple applications. Surmet has teamed with a major filled-resin products manufacturer who sees a high commercialization potential for our technology. This team-up will promote development of technology that is well suited for commercialization beyond Phase II.

36 MATERIALS SCIENCE↗

Automatic Energy Schemes for High Performance Applications

Although high-performance computing traditionally focuses on the efficient execution of large-scale applications, both energy and power have become critical concerns when approaching exascale. Drastic increases in the power consumption of supercomputers affect significantly their operating costs and failure rates. In modern microprocessor architectures, equipped with dynamic voltage and frequency scaling (DVFS) and CPU clock modulation (throttling), the power consumption may be controlled in software. Additionally, network interconnect, such as Infiniband, may be exploited to maximize energy savings while the application performance loss and frequency switching overheads must be carefully balanced. This work first studies two important collective communication operations, all-to-all and allgather and proposes energy saving strategies on the per-call basis. Next, it targets point-to-point communications to group them into phases and apply frequency scaling to them to save energy by exploiting the architectural and communication stalls. Finally, it proposes an automatic runtime system which combines both collective and point-to-point communications into phases, and applies throttling to them apart from DVFS to maximize energy savings. The experimental results are presented for NAS parallel benchmark problems as well as for the realistic parallel electronic structure calculations performed by the widely used quantum chemistry package GAMESS. Close to the maximum energy savings were obtained with a substantially low performance loss on the given platform.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Analysis of Debris Trajectories at the Scaled Wind Farm Technology (SWiFT) Facility

Sandia National Laboratories operates the Scaled Wind Farm Technology Facility (SWiFT) on behalf of the Department of Energy Wind and Water Power Technologies Office. An analysis was performed to evaluate the hazards associated with debris thrown from one of SWiFT’s operating wind turbines, assuming a catastrophic failure. A Monte Carlo analysis was conducted to assess the complex variable space associated with debris throw hazards that included wind speed, wind direction, azimuth and pitch angles of the blade, and percentage of the blade that was separated. In addition, a set of high fidelity explicit dynamic finite element simulations were performed to determine the threshold impact energy envelope for the turbine control building located on-site. Assuming that all of the layered, independent, passive and active engineered safety systems and administrative procedures failed (a 100% failure rate of the safety systems), the likelihood of the control building being struck was calculated to be less than 5/10,000 and ballistic simulations showed that the control building would not provide passive protection for the majority of impact scenarios. Although options exist to improve the ballistic resistance of the control building, the recommendation is not to pursue them because there is a low probability of strike and there is an equal likelihood personnel could be located at similar distances in other areas of the SWiFT facility which are not passively protected, while the turbines are operating. A fenced exclusion area has been created around the turbines which restricts access to the boundary of the 1/100 strike probability. The overall recommendation is to neither relocate nor improve passive protection of the control building as the turbine safety systems have been improved to have no less than two independent, redundant, high quality engineered safety systems. Considering this, in combination with a control building strike probability of less than 5/10,000, the overall probability of turbine debris striking the control building is less than 1/1,000,000.

17 WIND ENERGY↗

A Fault-oblivious Extreme-scale Execution Environment

Exascale computing systems will provide a thousand-fold increase in parallelism and a proportional increase in failure rate relative to today's machines. Systems software for exascale machines must provide the infrastructure to support existing applications while simultaneously enabling efficient execution of new programming models that naturally express dynamic, adaptive, irregular computation; coupled simulations; and massive data analysis in a highly unreliable hardware environment with billions of threads of execution. We propose a new approach to the data and work distribution model provided by system software based on the unifying formalism of an abstract file system. The proposed hierarchical data model provides simple, familiar visibility and access to data structures through the file system hierarchy, while providing fault tolerance through selective redundancy. The hierarchical task model features work queues whose form and organization are represented as file system objects. Data and work are both first class entities. By exposing the relationships between data and work to the runtime system, information is available to optimize execution time and provide fault tolerance. The data distribution scheme provides replication (where desirable and possible) for fault tolerance and efficiency, and it is hierarchical to make it possible to take advantage of locality. The user, tools, and applications, including legacy applications, can interface with the data, work queues, and one another through the abstract file model. This runtime environment will provide multiple interfaces to support traditional Message Passing Interface applications, languages developed under DARPA's High Productivity Computing Systems program, as well as other, experimental programming models. We will validate our runtime system with pilot codes on existing platforms and will use simulation to validate for exascale-class platforms. In this final report, we summarize research results from the work done at the Ohio State University towards the larger goals of the project listed above.

97 MATHEMATICS AND COMPUTING↗

Experimental assessment of fuel-cladding interactions

A range of fuel concepts designed to better tolerate accident scenarios and reactor transients are currently undergoing fundamental development at national laboratories as well as university and industrial partners. Pellet-clad mechanical and chemical interaction can be expected to affect fuel failure rates experienced during steady state operation, as well as dramatically impact the response of the fuel form under loss of coolant and other accident scenarios. The importance of this aspect of fuel design prompted research initiated by AFC in FY14 to begin exploratory efforts to characterize this phenomenon for candidate fuelcladding systems of immediate interest. Continued efforts in FY15 and FY17 aimed to better understand and simulate initial pellet-clad interaction with little-to-no pressure on the pellet-clad interface. Reported here are the results from 1000 h heat treatments at 400, 500, and 600°C of diffusion couples pairing UN with a FeCrAl alloy, SiC, and Zr-based cladding candidate sealed in evacuated quartz ampoules. No gross reactions were observed, though trace elemental contaminants were identified.

36 MATERIALS SCIENCE↗

Benchmark for Fuel Shuffling and Depletion for Pebble-Bed Reactors

Pebble bed reactors have specific operational characteristics when their fuel-cycle and fueling operations are considered. They are specifically distinguished by other type of nuclear reactor designs by their online fuel recycling scheme, where the fuel elements that have not yet reached discharge burnup can be reloaded and recycled continuously during normal operation. The fuel in a pebble bed reactor is not stationary and stochastically moves through the core once or several times during its lifetime, which allows them to operate without requiring a large excess reactivity hold for the burnup. However, this characteristic of pebble bed reactors introduces challenges in simulation, as each pebble can take many different trajectories through the core, its composition depends on the details of the irradiation history that is unique to its aggregated path through the core. For predicting the safety performance characteristics, such as source term, maximum fuel temperatures and fuel failure rates, etc., it is important to accurately incorporate the movement of pebbles through the core during their lifetime in a multi-physics simulation together with other phenomena. The equilibrium core analysis for pebble bed reactors are performed with multi-physics tools including fuel depletion in a multi pass reload coupled to the fuel movement. Currently, there are only a few legacy multi-physics simulation tools that can implement the pebble flow characteristics and perform equilibrium core analysis for pebble bed reactors. However, there are development efforts on-going under Department of Energy's Nuclear Energy Advanced Modelling and Simulation program and also in private industry for including these capabilities into their modelling and simulation tools. Any new development in the modelling and simulation tools needs to be validated by using tools such as experiments, analytical solutions or code-to-code benchmarks. In this work, a code-to-code benchmark for the equilibrium core analysis capability of pebble bed reactors was developed. Multiple cases were identified to capture different fuel cycle strategies that can be used in PBRs. The results of each case are presented in terms of overall equilibrium core characteristics: the discharge burnup; spatial burnup distribution; spatial isotopic distributions; axial and radial neutron flux distributions and power history of fuel elements per pass through the core for both a prototypical pebble bed High Temperature Gas-cooled Reactor and a prototypical pebble bed Fluoride-salt cooled High temperature Reactor.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Topology Optimization with a Manufacturability Objective

Part distortion and residual stress are critical factors for metal additive manufacturing (AM) because they can lead to high failure rates during both manufacturing and service. We present a topology optimization approach that incorporates a fast AM process simulation at each design iteration to provide predictions of manufacturing outcomes (i.e., residual stress, distortion, residual elastic energy) that can be optimized or constrained. The details of the approach and implementation are discussed, and an example design is presented that illustrates the efficacy of the method.

42 ENGINEERING↗

Operating Lifetime Study of Ultraviolet (UV) Light-Emitting Diode (LED) Products

Light-emitting diodes (LEDs) can emit radiation that spans the range from near infrared (IR) to all three bands of ultraviolet (UV) radiation: UV-A, UV-B, and UV-C. This report focuses on LEDs that emit in one of the three UV bands because they have the potential to displace lowpressure mercury vapor (LPMV) lamps in a variety of industrial processes, including ink and adhesive curing, medical procedures, and germicidal disinfection. However, before emerging UV LED technologies can displace LPMV lamps, the efficiency and reliability of these sources must meet the user’s expectations in each application. An earlier report focused on the construction and initial performance of commercial UV LED products in radiometric and current-voltage (IV) tests [1]. This report focuses on the long-term performance and reliability of the same set of commercial products. The intent of this report is to provide to the lighting industry a benchmark of the state of UV LEDs as of mid-2021 when these products were purchased. Understanding the failure modes and failure rates of UV LEDs is important in improving UV product reliability at the LED, lamp, and luminaire level and is critical to developing products with higher efficiency, lower carbon footprint, and significantly reduced environmental impact than LPMV lamps.

42 ENGINEERING↗

Enhanced Component Performance Study: Motor Driven Pumps 1998–2022

This report presents an enhanced performance evaluation of motor-driven pumps (MDPs) at U.S. commercial nuclear power plants. The data used in this study are based on the operating experience failure reports from calendar year 1998 through 2022 as reported in the Institute of Nuclear Power Operations (INPO) Industry Reporting and Information System (IRIS). The MDP failure modes considered for standby systems are fail to start (FTS), fail to run (FTR) for one hour of operation (FTR=1H), FTR after one hour of operation (FTR>1H), and for normally running systems FTS and FTR. An eight-hour unreliability estimate is also calculated and trended. The component reliability estimates and the reliability data are trended for the most recent 10-year period while yearly estimates for reliability are provided for the entire study period. The following increasing trends were identified for MDPs for the most recent 10-year period: (1) Standby MDP frequency of start demands (demands per reactor year), (2) Standby MDP frequency of FTR=1H hours (hours per reactor year), (3) Standby MDP frequency of FTR>1H hours, and (4) Normally running MDP frequency of run hours. The following decreasing trends were identified for MDPs for the most recent 10-year period: (1) Standby MDP FTR=1H failure probability, (2) Normally running MDP FTR failure rate, (3) Standby MDP unavailability, (4) Standby MDP total unreliability (8-hour mission), and (5) Normally running MDP total unreliability (8-hour mission).

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Instrumentation for the Investigation of Pitch Bearing Design and Reliability

Recently, there has been an increasing level of industry interest in pitch system and pitch bearing reliability. Pitch bearings are used in wind turbines to connect the blade root to the hub. Some populations of pitch bearings have demonstrated a 12% failure rate in 20 years. As rotor diameters continue to increase for tall land-based and offshore wind turbines, pitch bearings are becoming even larger in diameter, which can make them vulnerable to deflections and consequent stress concentrations. There is an increased need to more accurately study pitch bearing deformations, misalignment, load distributions, and contact stresses. A significant body of work has investigated fatigue lives and wear characteristics of pitch bearings on ground-based test rigs. NREL has also recently begun a research program related to pitch bearing reliability, recognizing its growing importance for wind turbines. The purpose of this paper is to describe a set of instrumentation that was recently installed on a 1.5 MW wind turbine at the NREL Flatirons Campus and provide an example data set. To the authors' knowledge, this will be the first publicly available pitch bearing data collection campaign on an operational wind turbine.

17 WIND ENERGY↗

Increasing Reliability and Safety of Hydrogen Components - Reliability Data Collection

Come learn about the new Hydrogen Component Reliability Database (HyCReD) and participate in discussions on hydrogen component reliability data collection, collaboration, and analysis. Funded by the U.S. Department of Energy's Office of Energy Efficiency and Renewable Energy under the Hydrogen and Fuel Cell Technologies Office, HyCReD is a collaborative project between the National Renewable Energy Laboratory, the University of Maryland, and hydrogen stakeholders to improve safety reliability for hydrogen facilities by integrating risk reduction methodologies and component reliability data taxonomies that support hydrogen infrastructure failure rate analysis.

component↗

Performance Test of Mini LVDT - ELVIS

The Institute for Energy (IFE) Technology has been a pioneer in the development of Linear Variable Differential Transformers (LVDTs) for in-pile testing, deploying over 2,200 units in various reactor environments with less than a 10% failure rate after five years of operation. This report focuses on the performance testing of IFE’s Mini LVDT, a compact sensor ideal for material test reactor experiments. The Mini LVDT, with a limited range of +/- 1.5 mm, offers excellent performance comparable to larger LVDTs, making it valuable in space-constrained applications. The development of an Enhanced Linear Variable Intrinsic Sensor (ELVIS) with internal temperature monitoring capabilities represents a significant advancement, addressing the critical need for real-time, accurate measurements in high-radiation and high-temperature environments. Two ELVIS prototypes were evaluated in terms of both temperature and displacement, showcasing promising results, though challenges with noise during temperature measurements were identified. This report summarizes the rigorous testing performed at Idaho National Laboratory (INL) and highlights the potential applications of Mini LVDTs and ELVIS in nuclear and other high-precision industries.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Evaluation of LLM-Generated Kokkos Code Using Compile-Time and Run-Time Testing

Due to the growing use of large language models (LLMs) by developers and researchers, it has become essential to reliably evaluate their ability to generate code that uses specialized libraries. We explore the use of compile-time and run-time evaluation of LLM-generated Kokkos code through extending the methods used by OpenAI with the HumanEval dataset. Our evaluation framework is based on the first 40 prompts from the Kokkos138 dataset. We start by discussing two different forms of LLM prompting, using entirely plain English or providing pseudocode for added context. These two methods are used to generate Kokkos code with the Llama-3.1-8B-Instruct and CodeQwen1.5-7B-Chat models. We found that both forms of prompting led to high failure rates and difficulties with reliably parsing LLM-generated code, while prompts with pseudocode for context generally led to improved results on more complicated tests.

97 MATHEMATICS AND COMPUTING↗