Search NASA⌕ Search

SEARCH · Search NASA

Results for “architecture optimization”

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 559 records · Page 31
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Shifting Between Compute and Memory Bounds: A Compression-Enabled Roofline Model

In the evolving landscape of high-performance computing, especially to fight the end of Moore’s Law and Dennard’s Scaling, the ability to shift between compute-bound and memory-bound states is critical for enhancing adaptability and flexibility to diverse system and domain-specific architectures. Such capability is vital for optimizing performance across distinguished hardware configurations, such as accelerators, memory hierarchies, and cache systems. Despite that ad hoc optimization techniques, such as compressed/approximate computation, have been enabled for compute-/data-intensive computing for improved performance in distinct hardware settings, there lacks an understanding of 1) the rational behind performance improvement; 2) capability of different optimizations; 3) what optimization to respond to specific computational and memory demands. This work proposes a compression-enabled roofline model to facilitate this adaptability with data compression techniques to balance and transform between computational and memory demands. This model enables applications to adjust in response to the specific strengths and limitations of the underlying hardware and system to optimize resource utilization. The effectiveness of this approach is demonstrated with matrix multiplication kernels on different input sizes, with turning on/off various compression techniques, including 1) low-precision floating point; 2) sparse matrix formulation; and 3) compressed arrays with ZFP. By reducing memory transfer volumes and cache misses and increasing data locality and computational intensity through compression, the specific roofline model can transform between compute and memory bounds to align more efficiently with system capabilities. This advancement not only improves overall performance but also maximizes adaptability in diverse computing environments.

Naraparaju, Ramasoumya [University of Washington]↗
DOE OSTI2017

Galactos: computing the anisotropic 3-point correlation function for 2 billion galaxies

The nature of dark energy and the complete theory of gravity are two central questions currently facing cosmology. A vital tool for addressing them is the 3-point correlation function (3PCF), which probes deviations from a spatially random distribution of galaxies. However, the 3PCF's formidable computational expense has prevented its application to astronomical surveys comprising millions to billions of galaxies. We present Galactos, a high-performance implementation of a novel, O( N 2 ) algorithm that uses a load-balanced k-d tree and spherical harmonic expansions to compute the anisotropic 3PCF. Our implementation is optimized for the Intel Xeon Phi architecture, exploiting SIMD parallelism, instruction and thread concurrency, and significant L1 and L2 cache reuse, reaching 39% of peak performance on a single node. Galactos scales to the full Cori system, achieving 9.8 PF (peak) and 5.06 PF (sustained) across 9636 nodes, making the 3PCF easily computable for all galaxies in the observable universe.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

Ultrafast Error-bounded Lossy Compression for Scientific Datasets

Today's scientific high-performance computing applications and advanced instruments are producing vast volumes of data across a wide range of domains, which impose a serious burden on data transfer and storage. Error-bounded lossy compression has been developed and widely used in the scientific community because it not only can significantly reduce the data volumes but also can strictly control the data distortion based on the user-specified error bound. Existing lossy compressors, however, cannot offer ultrafast compression speed, which is highly demanded by numerous applications or use cases (such as in-memory compression and online instrument data compression). In this paper, we propose a novel ultrafast error-bounded lossy compressor that can obtain fairly high compression performance on both CPUs and GPUs and with reasonably high compression ratios. The key contributions are threefold. (1) We propose a generic error-bounded lossy compression framework---called SZx---that achieves ultrafast performance through its novel design comprising only lightweight operations such as bitwise and addition/subtraction operations, while still keeping a high compression ratio. (2) We implement SZx on both CPUs and GPUs and optimize the performance according to their architectures. (3) We perform a comprehensive evaluation with six real-world production-level scientific datasets on both CPUs and GPUs. Experiments show that SZx is 2~16x faster than the second-fastest existing error-bounded lossy compressor (either SZ or ZFP) on CPUs and GPUs, with respect to both compression and decompression.

Yu, Xiaodong↗
DOE OSTI2016

Flexible Computer Science Infrastructure (FleCSI)

FleCSI is a compile-time configurable framework designed to support multi-physics application development. As such, FleCSI attempts to provide a very general set of infrastructure design patterns that can be specialized and extended to suit the needs of a broad variety of solver and data requirements. Current support includes multi-dimensional mesh topology, mesh geometry, and mesh adjacency information, n-dimensional hashed-tree data structures, graph partitioning interfaces, and dependency closures. FleCSI also introduces a functional programming model with control, execution, and data abstractions that are consistent with both MPI and state-of-the-art task-based runtimes such as Legion and Charm++. The FleCSI abstraction layer provides the developer with insulation from the underlying runtime, while allowing support for multiple runtime systems, including conventional models like asynchronous MPI. The intent is to give developers a concrete set of user-friendly programming tools that can be used now, while allowing flexibility in choosing runtime implementations and optimizations that can be applied to architectures and runtimes that arise in the future. The control and execution models in FleCSI also provide formal nomenclature for describing poorly understood concepts like kernels and tasks.

Moss, Nicholas↗
DOE OSTI2026

BSMV-mediated genome editing exhibits host-specific heritability: germline transmission in barley and somatic edits in Nicotiana benthamiana

Plant RNA virus–mediated guide RNA (gRNA) delivery represents a transformative advance in genome editing technologies. Unlike conventional transformation methods that rely on labor-intensive tissue culture and regeneration for each individual gRNA delivery, viral vectors can rapidly and systemically transmit gRNAs into pre-established Cas-expressing plants, providing an accelerated route for functional genomics and trait discovery directly in planta . However, key design parameters, including subgenomic promoter choice, transcript architecture, and their effects on viral fitness and editing outcomes, remain to be elucidated for most viral platforms. We developed five Barley stripe mosaic virus (BSMV) vectors, each with distinct subgenomic promoter elements to drive single gRNA expression. These were initially evaluated in Cas9-expressing transgenic Nicotiana benthamiana plants targeting the Phytoene desaturase ( PDS ) gene to compare their editing efficiencies. Single gRNAs expressed under the duplicated γb subgenomic promoter or when fused directly to the γb genome achieved the highest mutation frequencies (up to 90% at 60 days post-inoculation), whereas β1- and β2-driven sgRNAs produced delayed and reduced editing. Thus, promoter selection critically determines gRNA accumulation and the efficacy of BSMV-mediated genome editing. The top-performing design was then applied to Cas9-expressing barley ( Hordeum vulgare ) targeting HvCMF7 (conferring green-white variegation) and HvGW2.1 (impacts grain width and weight). BSMV spread systemically throughout barley, inducing somatic and heritable mutations at frequencies up to 100%, with virus-free edited progeny. In contrast, despite robust somatic editing in N. benthamiana, no heritable mutations were detected indicating species-dependent limitations in germline transmission. Our systematic comparison of subgenomic promoter architectures establishes clear design principles for optimizing viral vector–mediated delivery. Promoter choice and transcript structure critically shape editing efficiency and viral stability. The host-specific boundary for germline editing, defined by efficient heritable editing in barley but not N. benthamiana , highlights where BSMV offers advantages and where alternative vectors or hybrid strategies are required, guiding rational platform selection for diverse crop species and applications. Collectively, these findings establish BSMV as a promising next-generation vector for rapid, tissue culture–free, and transformation-independent genome editing in cereals and other recalcitrant monocots.

barley↗
DOE OSTI2023

High-bandwidth image-based predictive laser stabilization via optimized Fourier filters

Controlling the delivery of kHz-class pulsed lasers is of interest in a variety of industrial and scientific applications, from next-generation laser-plasma acceleration to laser-based x-ray emission and high-precision manufacturing. The transverse position of the laser pulse train on the application target is often subject to fluctuations by external drivers (e.g., room cooling and heating systems, motorized optics stages and mounts, vacuum systems, chillers, and/or ground vibrations). For typical situations where the disturbance spectrum exhibits discrete peaks on top of a broad-bandwidth lower-frequency background, traditional PID (proportional-integral-derivative) controllers may struggle, since as a general rule PID controllers can be used to suppress vibrations up to only about 5%–10% of the sampling frequency. Here, a predictive feed-forward algorithm is presented that significantly enhances the stabilization bandwidth in such laser systems (up to the Nyquist limit at half the sampling frequency) by online identification and filtering of one or a few discrete frequencies using optimized Fourier filters. Furthermore, the system architecture demonstrated here uses off-the-shelf CMOS cameras and piezo-electric actuated mirrors connected to a standard PC to process the alignment images and implement the algorithm. To avoid high-end, high-cost components, a machine-learning-based model of the piezo mirror’s dynamics was integrated into the system, which enables high-precision positioning by compensating for hysteresis and other hardware-induced effects. A successful demonstration of the method was performed on a 1 kHz laser pulse train, where externally-induced vibrations of up to 400 Hz were attenuated by a factor of five, far exceeding what could be done with a standard PID scheme.

Natal, Joseph↗
DOE OSTI2016

Contingency Contractor Optimization Phase 3 Sustainment Software Design Document - Contingency Contractor Optimization Tool - Prototype

This document describes the final software design of the Contingency Contractor Optimization Tool - Prototype. Its purpose is to provide the overall architecture of the software and the logic behind this architecture. Documentation for the individual classes is provided in the application Javadoc. The Contingency Contractor Optimization project is intended to address Department of Defense mandates by delivering a centralized strategic planning tool that allows senior decision makers to quickly and accurately assess the impacts, risks, and mitigation strategies associated with utilizing contract support. The Contingency Contractor Optimization Tool - Prototype was developed in Phase 3 of the OSD ATL Contingency Contractor Optimization project to support strategic planning for contingency contractors. The planning tool uses a model to optimize the Total Force mix by minimizing the combined total costs for selected mission scenarios. The model optimizes the match of personnel types (military, DoD civilian, and contractors) and capabilities to meet mission requirements as effectively as possible, based on risk, cost, and other requirements.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

ATDM LANL FleCSI: Topology and Execution Framework

FleCSI is a compile-time configurable C++ framework designed to support multi-physics application development. As such, FleCSI attempts to provide a very general set of infrastructure design patterns that can be specialized and extended to suit the needs of a broad variety of solver and data requirements. This means that FleCSI is potentially useful to many different ECP projects. Current support includes multidimensional mesh topology, mesh geometry, and mesh adjacency information, n-dimensional hashed-tree data structures, graph partitioning interfaces, and dependency closures (to identify data dependencies between distributed-memory address spaces). FleCSI introduces a functional programming model with control, execution, and data abstractions that are consistent with state-of-the-art task-based runtimes such as Legion and Charm++. The model also provides support for fine-grained, data-parallel execution with backend support for runtimes such as OpenMP and C++17. The FleCSI abstraction layer provides the developer with insulation from the underlying runtimes, while allowing support for multiple runtime systems, including conventional models like asynchronous MPI. The intent is to give developers a concrete set of user-friendly programming tools that can be used now, while allowing flexibility in choosing runtime implementations and optimizations that can be applied to architectures and runtimes that arise in the future. This project is essential to the ECP Ristra Next-Generation Code project, part of ASC ATDM, because it provides a hierarchically parallel programming model that is consistent with the design of modern system architectures, but which allows for the straightforward expression of algorithmic parallelism in a portably performant manner.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

Compositionally Graded Alloy Quantum Dot Phosphors for Solid State Lighting. Final Report

To push the boundaries of efficiency and cost in solid state lighting (SSL), narrow band red emitting down conversion phosphors must be improved. These materials are critical to both the color rendering ability and optical output of phosphor converted light emitting diodes (LEDs). Improvements in these performance parameters will have a significant impact on the adoption of SSL in retail, office, and home lighting applications. Quantum Dots (QDs) are widely acknowledged as a cost-effective front-runner among developing downconverting technologies because they emit narrow band luminescence that is tunable across the visible spectrum. Current syntheses of quantum dots, however, utilize expensive and toxic reagents and often involve several time-intensive purification steps. To address this, we have introduced a novel, cost-effective synthetic scheme, whereby the final quantum dot structures can be accessed with fewer synthetic steps using a library of inexpensive, easily-synthesized precursors. With the help of automated high throughput synthesis robotics, Columbia, LBNL, and OSRAM have optimized red and amber CdS/CdSe/CdS QD architectures for SSL. QDs with improved shelling layers have been identified that meet the industry state of the art performance on chip under high humidity accelerated aging tests. This is remarkable considering the time frame of this project compared to the time and resources allocated for the development of the industrial standard. Furthermore, although particle stability to optical charging is often stressed in improving QLED performance, we have identified effective ZnS particle over-coating as the most important factor determining particle survivability and device performance in down-converting LEDs.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Exago TM Users Manual: Version 1.0

The Exascale Grid Optimization (ExaGOTM) toolkit is an open source package for solving large-scale power grid optimization problems on parallel and distributed architectures, particularly targeted for exascale machines with heteregenous architectures (GPU). This manual is a guide to ExaGO's working including installation, formulation, and usage.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2026

From IMT Device Measurements to Network-Level Consequences: When Learning Suppresses Beyond-LIF Neuron Dynamics

Emerging neuromorphic devices such as insulator--metal transition (IMT) devices exhibit complex temporal dynamics, including slow internal state memory, hysteresis, and burst-like firing, which are poorly captured by conventional leaky integrate-and-fire (LIF) neurons. However, it remains unclear when such dynamics influence learning and inference at the network level, particularly under commonly used unsupervised plasticity rules. We present a controlled, full-stack co-design study spanning experimental characterization of individual IMT devices, compact neuron model development, and large-scale spiking network simulations with identical architectures and learning rules. Rather than optimizing benchmark accuracy, our goal is to diagnose when neuron-level dynamics survive learning and competition, and when they are suppressed, to inform the co-design of devices, networks, and learning rules that can exploit beyond-LIF complexity.

42 ENGINEERING↗
NASA NTRS1989

Approximation algorithms for planning and control

A control system operating in a complex environment will encounter a variety of different situations, with varying amounts of time available to respond to critical events. Ideally, such a control system will do the best possible with the time available. In other words, its responses should approximate those that would result from having unlimited time for computation, where the degree of the approximation depends on the amount of time it actually has. There exist approximation algorithms for a wide variety of problems. Unfortunately, the solution to any reasonably complex control problem will require solving several computationally intensive problems. Algorithms for successive approximation are a subclass of the class of anytime algorithms, algorithms that return answers for any amount of computation time, where the answers improve as more time is allotted. An architecture is described for allocating computation time to a set of anytime algorithms, based on expectations regarding the value of the answers they return. The architecture described is quite general, producing optimal schedules for a set of algorithms under widely varying conditions.

Boddy, Mark↗
NASA NTRS1992

Frequency allocations for a new satellite service - Digital audio broadcasting

The allocation in the range 500-3000 MHz for digital audio broadcasting (DAB) is described in terms of key issues such as the transmission-system architectures. Attention is given to the optimal amount of spectrum for allocation and the technological considerations relevant to downlink bands for satellite and terrestrial transmissions. Proposals for DAB allocations are compared, and reference is made to factors impinging on the provision of ground/satellite feeder links. The allocation proposals describe the implementation of 50-60-MHz bandwidths for broadcasting in the ranges near 800 MHz, below 1525 MHz, near 2350 MHz, and near 2600 MHz. Three specific proposals are examined in terms of characteristics such as service areas, coverage/beam, channels/satellite beam, and FCC license status. Several existing problems are identified including existing services crowded with systems, the need for new bands in the 1000-3000-MHz range, and variations in the nature and intensity of implementations of existing allocations that vary from country to country.

Reinhart, Edward E.↗
NASA NTRS1997

OPAD-EDIFIS Real-Time Processing

The Optical Plume Anomaly Detection (OPAD) detects engine hardware degradation of flight vehicles through identification and quantification of elemental species found in the plume by analyzing the plume emission spectra in a real-time mode. Real-time performance of OPAD relies on extensive software which must report metal amounts in the plume faster than once every 0.5 sec. OPAD software previously written by NASA scientists performed most necessary functions at speeds which were far below what is needed for real-time operation. The research presented in this report improved the execution speed of the software by optimizing the code without changing the algorithms and converting it into a parallelized form which is executed in a shared-memory multiprocessor system. The resulting code was subjected to extensive timing analysis. The report also provides suggestions for further performance improvement by (1) identifying areas of algorithm optimization, (2) recommending commercially available multiprocessor architectures and operating systems to support real-time execution and (3) presenting an initial study of fault-tolerance requirements.

Katsinis, Constantine↗
NASA NTRS2000

The Instrument Synthesis for Analysis Laboratory at the Goddard Space Flight Center

The objectives of the Instrument Synthesis and Analysis laboratory (ISAL) are to: 1) Provide a rapid and sustainable instrument development environment with clear efficient processes and tools, re-usable models and skilled engineers. 2) Provide a capability for quick and efficient trade studies of instrument architectures and concepts. 3) Streamline and optimize instrument systems design for the entire life cycle, including cost and technology assessment. 4) Provide detailed multi-disciplinary modeling and analysis for phase C/D design validation and for operational investigations.

Wood, H. John↗
NASA NTRS1994

High Speed Computing, LANs, and WAMs

Optical fiber networks may one day offer potential capacities exceeding 10 terabits/sec. This paper describes present gigabit network techniques for distributed computing as illustrated by the CASA gigabit testbed, and then explores future all-optic network architectures that offer increased capacity, more optimized level of service for a given application, high fault tolerance, and dynamic reconfigurability.

all-optics networks terabit fiber optic networks s↗
NASA NTRS2008

Range Imaging without Moving Parts

Range-imaging instruments of a type now under development are intended to generate the equivalent of three-dimensional images from measurements of the round-trip times of flight of laser pulses along known directions. These instruments could also provide information on characteristics of targets, including roughnesses and reflectivities of surfaces and optical densities of such semi-solid objects as trees and clouds. Unlike in prior range-imaging instruments based on times of flight along known directions, there would be no moving parts; aiming of the laser beams along the known directions would not be accomplished by mechanical scanning of mirrors, prisms, or other optical components. Instead, aiming would be accomplished by using solid-state devices to switch input and output beams along different fiber-optic paths. Because of the lack of moving parts, these instruments could be extraordinarily reliable, rugged, and long-lasting. An instrument of this type would include an optical transmitter that would send out a laser pulse along a chosen direction to a target. An optical receiver coaligned with the transmitter would measure the temporally varying intensity of laser light reflected from the target to determine the distance and surface characteristics of the target. The transmitter would be a combination of devices for generating precise directional laser illumination. It would include a pulsed laser, the output of which would be coupled into a fiber-optic cable with a fan-out and solid-state optical switches that would enable switching of the laser beam onto one or more optical fibers terminated at known locations in an array on a face at the focal plane of a telescope. The array would be imaged by the telescope onto the target space. The receiver optical system could share the aforementioned telescope with the transmitter or could include a separate telescope aimed in the same direction as that of the transmitting telescope. In either case, light reflected from the target would be focused by the receiver optical system onto an array of optical fibers matching the array in the transmitter. These optical fibers would couple the received light to one or more photodetector( s). Optionally, the receiver could include solid-state optical switches for choosing which optical fiber(s) would couple light to the photodetector(s). This instrument architecture is flexible and can be optimized for a wide variety of applications and levels of performance. For example, it is scalable to any number of pixels and pixel resolutions and is compatible with a variety of ranging and photodetection methodologies, including, for example, ranging by use of modulated (including pulsed and encoded) light signals. The use of fixed arrays of optical fibers to generate controlled illumination patterns would eliminate the mechanical complexity and much of the bulk of optomechanical scanning assemblies. Furthermore, digital control of the selection of the fiber-optic pathways for the transmitted beams could afford capabilities not seen in previous three-dimensional range-imaging systems. Instruments of this type could be specialized for use as, for example, proximity detectors, three-dimensional robotic vision systems, airborne terrain-mapping systems, and inspection systems.

Blair, J. Bryan↗
NASA NTRS2006

Automated Synthesis of Architecture of Avionic Systems

The Architecture Synthesis Tool (AST) is software that automatically synthesizes software and hardware architectures of avionic systems. The AST is expected to be most helpful during initial formulation of an avionic-system design, when system requirements change frequently and manual modification of architecture is time-consuming and susceptible to error. The AST comprises two parts: (1) an architecture generator, which utilizes a genetic algorithm to create a multitude of architectures; and (2) a functionality evaluator, which analyzes the architectures for viability, rejecting most of the non-viable ones. The functionality evaluator generates and uses a viability tree a hierarchy representing functions and components that perform the functions such that the system as a whole performs system-level functions representing the requirements for the system as specified by a user. Architectures that survive the functionality evaluator are further evaluated by the selection process of the genetic algorithm. Architectures found to be most promising to satisfy the user s requirements and to perform optimally are selected as parents to the next generation of architectures. The foregoing process is iterated as many times as the user desires. The final output is one or a few viable architectures that satisfy the user s requirements.

Chau, Savio↗