SEARCH · Search NASA
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.
Under IER-206, the hardware for the Burnup Credit Critical Experiment will be removed from storage and placed in operation. This is the hardware we used in 2002 to perform the experiments that eventually documented in the International Handbook of Evaluated Criticality Safety Benchmark Experiments as LEU-COMP-THERM-079.
This report summarizes a set of figures of merit of interest in monitoring the hardware and hardware usage in a Sandia high performance computing (HPC) center. These figures are computable from high frequency monitoring data and other non-metric data and may aid administrators and customer support personnel in their decision processes. The figures are derived from interviews of the HPC center staff. The figures are in many cases simplistic data reductions, but they are our initial targets in creating dashboards that turn voluminous monitoring data into actionable information. Because simplistic reductions may obscure as well as reveal the situation under study, we also document the necessary 'drill-down' and %60exploration' views needed to make the data better understood quickly. These figures of merit may be compared to dashboarding tools documented by other HPC centers.
Graph Neural Networks (GNNs) have garnered a lot of recent interest because of their success in learning representations from graph-structured data across several critical applications in cloud and HPC. Owing to their unique compute and memory characteristics that come from an interplay between dense and sparse phases of computations, the emergence of reconfigurable dataflow (aka spatial) accelerators offers promise for acceleration by mapping optimized dataflows (i.e., computation order and parallelism) for both phases. The goal of this work is to characterize and understand the design-space of dataflow choices for running GNNs on spatial accelerators in order for the compilers to optimize the dataflow based on the workload. Specifically, we propose a taxonomy to describe all possible choices for mapping the dense and sparse phases of GNNs spatially and temporally over a spatial accelerator, capturing both the intra-phase dataflow and the inter-phase (pipelined) dataflow. Using this taxonomy, we do deep-dives into the cost and benefits of several dataflows and perform case studies on implications of hardware parameters for dataflows and value of flexibility to support pipelined execution.
The Department of Energy (DOE)’s Advanced Scientific Computing Research (ASCR) program has developed hundreds of software tools for high performance computing (HPC). Although many HPC hardware advances have permeated the market, such as CPUs, GPUs, TPUs and ASICs, the software developed by DOE for different HPC applications has not necessarily been mined. The DOE is interested in making its HPC software portfolio available to the public in order to maximize the value of what has been developed.
Using a newly developed coupling of the ElectroMagnetic Plasma In Realistic Environments (EMPIRE) code with the Integrated Tiger Series (ITS) code, radiation environment calculations have been performed. The effort was completed as part of the Saturn Recapitalization (Recap) program that represents activities to upgrade and modernize the Saturn accelerator facility. The radiation environment calculations performed provide baseline results with current or planned hardware in the facility. As facility design changes are proposed and implemented as part of Saturn Recap, calculations of the radiation environment will be performed to understand how the changes impact the output of the Saturn accelerator.
Explore the source record for details and available documents.
PV panels are subjected to wind loads during normal outdoor operation, where th wind speed, wind direction, panel angle, and array layout play a large role in the overall loading magnitude. For floating PV systems, where panels are installed on floating rafts, these forces can lead to a dynamic displacement of the raft and mounted PV hardware. The Contractor and Participant will perform fluid-structure interaction simulations of this phenomenon for a variety of wind speeds, directions, and panel angles to characterize these forces and help design mooring/tethering lines to resist and anchor raft movement.
A technical history and managerial critique of the MSFC role in the Skylab program is presented. The George C. Marshall Space Flight Center had primary hardware development responsibility for the Saturn Workshop Modules and many of the designated experiments in addition to the system integration responsibility for the entire Skylab Orbital Cluster. The report also includes recommendations and conclusions applicable to hardware design, test program philosophy and performance, and program management techniques with potential application to future programs.
Software and hardware requirements for implementing hardware self tests are presented in support of the development of training and procedures development simulators for the space shuttle program. Self test techniques for simulation hardware and the validation of simulation performance are stipulated. The requirements of an integrated simulator self system are analyzed. Readiness tests, fault isolation tests, and incipient fault detection tests are covered.
The conceptual design of a suspension system for the Floated Pallet and a retention system to support the pallet during launch, maneuvering and descent are described. In addition, a control moment gyroscope support equipment configuration study including installation and an experiment mount and erection study was performed. The impact of these separate hardware studies on the existing pallet configuration was assessed and a modified pallet common module defined.
The development of a high temperature solar thermal receiver is described. A prototype receiver and associated test support (auxiliary) hardware was fabricated. Shakedown and initial performance tests of the prototype receiver were performed. Maximum outlet temperatures of 1600 F were achieved at 100% solar (70-75 kW) input power with 900 F inlet temperatures and a subsequent testing was concluded by a 2550 F outlet run. The window retaining assembly was modified to improve its tolerance for thermal distortion of the flanges. It is shown that cost effective receiver designs can be implemented within the framework of present materials technology.
A conceptual framework for treating software quality from a total system perspective is developed. Examples are given to show how system quality objectives may be allocated to hardware and software; to illustrate trades among quality factors, both hardware and software, to achieve system performance objectives; and to illustrate the impact of certain design choices on software functionality.
The wrap-rib deployable antenna concept development is based on a combination of hardware development and testing along with extensive supporting analysis. The proof-of-concept hardware models are large in size so they will address the same basic problems associated with the design fabrication, assembly and test as the full-scale systems which were selected to be 100 meters at the beginning of the program. The hardware evaluation program consists of functional performance tests, design verification tests and analytical model verification tests. Functional testing consists of kinematic deployment, mesh management and verification of mechanical packaging efficiencies. Design verification consists of rib contour precision measurement, rib cross-section variation evaluation, rib materials characterizations and manufacturing imperfections assessment. Analytical model verification and refinement include mesh stiffness measurement, rib static and dynamic testing, mass measurement, and rib cross-section characterization. This concept was considered for a number of potential applications that include mobile communications, VLBI, and aircraft surveillance. In fact, baseline system configurations were developed by JPL, using the appropriate wrap-rib antenna, for all three classes of applications.
Data compression techniques are presented. A description of these techniques is provided along with a performance evaluation. The complexity of the hardware resulting from their implementation is also addressed. The compression effect on channel distortion and the applicability of these algorithms to real-time processing are presented. Also included is a proposed new direction for an adaptive compression technique for real-time processing.
To demonstrate the high reliability of fault-tolerant computing systems, experimental testing can be performed by injecting faults into their hardware and observing the times needed to detect and isolate failed components and reconfigure the good components. Coverage, a parameter measuring continued system success, is the probability that recovery occurs before other, perhaps catastrophic, component failures occur. Methods of estimating coverage are given for the case of randomly selecting a subset of the pins or chips for testing. Pin-level samples of times to recovery are treated by the Type I censoring model often used in life testing.
The normalized standard deviation, Kp, of radar backscatter measurements using digital Doppler processors in spaceborne scatterometers is derived. The Kp expression for analog Doppler filter processors, such as that used in the Seasat scatterometer is shown to be a special case of the derived Kp expression. A connection to Welch's (1967) power spectrum estimation rsults is also made. Tradeoff studies in digital filter design such as hardware complexity, computational speed, and system performance can be performed based on this Kp expression. A current application in the design of the NASA scatterometer to be flown in 1990 is discussed. This derivation should be useful for system design and analysis of other radar remote-sensing instruments.
Spacecraft telemetry systems have evolved from simple hardware devices to complex computer applications performing data acquisition and formatting tasks. This paper reviews the role of spacecraft computers in performing telemetry functions and examines computer-based telemetry systems being considered for use on the NASA Space Station.
Telemetry-data-acquisition unit adjusted remotely to produce changes in sampling rate, sampling channels, measurement scale, and output-bias level. Functional configuration adapted to changing conditions or new requirements by distant operator over telemetry link. Reconfiguration done in real time, without removing equipment from service. Bus-interface unit accepts reprogramming commands and translates them for low-rate adaptive multiplexer. Reprogrammable equipment reduces need for spare parts, since not necessary to stock variety of hardware with fixed characteristics. Adaptive multiplexer performs well in tests, amplitude errors less than 0.5 percent, distortion less than 0.25 percent, and crosstalk and common-mode rejection indiscernible.