Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “high level application”

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 1,207 records · Page 67

A Synthesis Methodology for Intelligent Memory Interfaces in Accelerator Systems

Domain-specific systems improve the performance of a specific set of applications compared to general-purpose processing systems by deploying custom hardware accelerators. These hardware accelerators are generated using high-level synthesis (HLS) tools. The HLS tools enable a comprehensive design space exploration to optimize the compute performance of the generated accelerators. However, they often ignore the challenges of implementing the accelerators in a system-on-chip, particularly how the accelerators access memory. Our work introduces a buffering system design that improves accelerators' memory accesses by intelligently employing burst transactions to prefetch useful data from external memory to on-chip local buffers. Our design is dynamic, parametric, and transparent to the accelerators generated by HLS tools. We derive the buffering system parameters using appropriate compiler-based analysis passes and memory channel latency constraints. The proposed buffering system design results in, on average, 8.8x performance improvements while lowering memory channel utilization on average by 53.2% for a set of PolyBench kernels.

Limaye, Ankur M. (ORCID:0000000194062584)↗

hls4ml: A Flexible, Open-Source Platform for Deep Learning Acceleration on Reconfigurable Hardware

We present hls4ml, a free and open-source platform that translates machine learning (ML) models from modern deep learning frameworks into high-level synthesis (HLS) code that can be integrated into full designs for field-programmable gate arrays (FPGAs) or application-specific integrated circuits (ASICs). With its flexible and modular design, hls4ml supports a large number of deep learning frameworks and can target HLS compilers from several vendors, including Vitis HLS, Intel oneAPI and Catapult HLS. Together with a wider eco-system for software-hardware co-design, hls4ml has enabled the acceleration of ML inference in a wide range of commercial and scientific applications where low latency, resource usage, and power consumption are critical. In this paper, we describe the structure and functionality of the hls4ml platform. The overarching design considerations for the generated HLS code are discussed, together with selected performance results.

Schulte, Jan-Frederik [Purdue U.] (ORCID:000000034↗

Effect of 1-hydroxyethane-1,1-diphosphonic acid (HEDPA) on Partitioning of Np and Pu to Synthetic Boehmite

The effect of 1-hydroxyethane-1,1-diphosphonic acid (HEDPA) on sorption of Np(V) and Pu(V) to synthetic boehmite ({gamma}-AlOOH) was examined a function of time and pH (between 4 to 11). Sorption of both elements in boehmite suspensions (1 M NaCl, 600 mg L{sup -1} boehmite) increased with increasing pH. Sorption edges for neptunium and plutonium occurred at approximately pH 8.0 and 6.6, respectively. After steady state partitioning was reached, HEDPA was added to the neptunium-boehmite and plutonium-boehmite suspensions. Neptunium and plutonium partitioning appears to be primarily affected by the formation of soluble Np:HEDPA and Pu:HEDPA complexes, the dissolution of boehmite promoted by HEDPA, and the precipitation of Np:HEDPA and Pu:HEDPA colloids. The results are discussed in terms of applicability of HEDPA-promoted dissolution as a waste reduction method in the treatment of sludge phases contained within high-level nuclear waste storage tanks.

54 ENVIRONMENTAL SCIENCES↗

The Development of a Robust Accelerometer-Based Start of Combustion Sensing System

Polymer nanofibers are nanoscale materials whose properties can be adjusted to provide desirable light management performance for high efficiency solid-state lighting luminaires. The polymeric nanofibers at the core of this project have diameters on the order of 100 to 1000 nm and a length of more than 1 cm. By controlling fiber diameter, fiber packing, and fiber morphology, a low cost, high performance optical material can be fabricated. This report describes the fabrication of these nanofiber structures and their uses and benefits in solid-state lighting application. When used in solid state lighting devices, nanofibers can take the form of either diffuse reflectors or photoluminescent materials. Nanofiber reflectors (NFR) were developed which displayed high diffuse reflectance with reflectance values in excess of 0.90. In contrast, traditional reflector materials such as aluminum and paint typically possess reflectance values below 0.80 and absorb a larger fraction of light, reducing luminaire output efficiency. The incorporation of the NFR technology into reflectors, troffers, and beam formers present in SSL luminaires provides better reflectance and lower light loss than is possible with conventional materials. Photoluminescent nanofibers (PLN) can be formed by combining nanofibers with photoluminescent materials such as phosphors and quantum dots (QD). Forming the PLN with the proper combination of green and red luminescent materials and exciting the nanocomposite with a blue light emitting diode (LED) has been demonstrated to produce high efficiency (> 55 lumens per watt) white light with excellent color rendering properties. The incorporation of QDs in the PLN is particularly advantageous in that this approach enables the correction of any color deficiencies in the light source without creating unnecessary radiation in the near infrared. Cost models developed during this project have demonstrated that both the NFR and PLN materials can be mass produced at a manufacturing cost of less than $5 per square foot, making it commercially attractive. To capitalize on the benefits of nanofiber technologies in solid-state lighting, several new remote phosphor reflector configurations were developed in the project. When combined with these unique lighting designs, nanofibers have a number of demonstrated benefits in lighting devices including: (1) Providing high quantum efficiency down-conversion of LED wavelengths to produce full spectrum white light; (2) Enabling tunable device structures that achieve colors ranging from warm white to cool white with high CRIs; (3) Supplying mass producible, cost-effective solutions for diffuse, high reflectance light management across the visible spectrum; (4) Facilitating remote phosphor luminaire designs that increase the lifetime and performance of luminescent materials; and (5) Providing conformability to geometries imposed by the light fixture enabling new lighting designs. This report provides a review of the activities conducted during this project and the major advances that have been made during this work in the field of nanoscale materials for solid-state lighting. Section 1 provides background on the nanofiber technologies and patent-pending luminaire structures developed during this project. Section 2 provides a detailed discussion of the benefits of the technologies developed during this work. Section 3 compares the execution of this project with the original proposal. A high level summary of the findings from each Task is also provided in the section. Section 4 lists the products (i.e., patent applications, presentations, publications, collaborations) that have been developed during this project. More details on work conducted during Budget Periods 1-3 can be found in Appendices A-C. The technologies developed during this program have significant commercial potential, and there has already been strong interest in these breakthroughs from the lighting community. It is anticipated that these technologies will begin appearing in commercial products in roughly 5 years to provide significant energy savings for the United States.

77 NANOSCIENCE AND NANOTECHNOLOGY↗

ConnectX2 In niBand Management Queues: New support for Network Of oaded

This paper introduces the newly developed InfiniBand (IB) Management Queue capability, used by the Host Channel Adapter (HCA) to manage network task data flow dependancies, and progress the communications associated with such flows. These tasks include sends, receives, and the newly supported wait task, and are scheduled by the HCA based on a data dependency description provided by the user. This functionality is supported by the ConnectX-2 HCA, and provides the means for delegating collective communication management and progress to the HCA, also known as collective communication offload. This provides a means for overlapping collective communications managed by the HCA and computation on the Central Processing Unit (CPU), thus making it possible to reduce the impact of system noise on parallel applications using collective operations. This paper further describes how this new capability can be used to implement scalable Message Passing Interface (MPI) collective operations, describing the high level details of how this new capability is used to implement the MPI Barrier collective operation, focusing on the latency sensitive performance aspects of this new capability. This paper concludes with small scale benchmark experiments comparing implementations of the barrier collective operation, using the new network offload capabilities, with established point-to-point based implementations of these same algorithms, which manage the data flow using the central processing unit. These early results demonstrate the promise this new capability provides to improve the scalability of high-performance applications using collective communications. The latency of the HCA based implementation of the barrier is similar to that of the best performing point-to-point based implementation managed by the central processing unit, starting to outperform these as the number of processes involved in the collective operation increases.

99 GENERAL AND MISCELLANEOUS↗

A very high speed lossless compression/decompression chip set

A chip is described that will perform lossless compression and decompression using the Rice Algorithm. The chip set is designed to compress and decompress source data in real time for many applications. The encoder is designed to code at 20 M samples/second at MIL specifications. That corresponds to 280 Mbits/second at maximum quantization or approximately 500 Mbits/second under nominal conditions. The decoder is designed to decode at 10 M samples/second at industrial specifications. A wide range of quantization levels is allowed (4...14 bits) and both nearest neighbor prediction and external prediction are supported. When the pre and post processors are bypassed, the chip set performs high speed entropy coding and decoding. This frees the chip set from being tied to one modeling technique or specific application. Both the encoder and decoder are being fabricated in a 1.0 micron CMOS process that has been tested to survive 1 megarad of total radiation dosage. The CMOS chips are small, only 5 mm on a side, and both are estimated to consume less than 1/4 of a Watt of power while operating at maximum frequency.

Venbrux, Jack↗

An FDMA system concept for 30/20 GHz high capacity domestic satellite service

The paper summarizes a feasibility study of a multibeam FDMA satellite system operating in the 30/20 GHz band. The system must accommodate a very high volume of traffic within the restrictions of a 5 kW solar cell array and a 2.5 GHz bandwidth. Multibeam satellite operation reduces the DC power demand and allows reuse of the available bandwidth. Interferences among the beams are brought to acceptable levels by appropriate frequency assignments. A transponder design is presented; it is greatly simplified by the application of a regional concept. System analysis shows that MSK modulation is appropriate for a high-capacity system because it conserves the frequency spectrum. Rain attenuation, a serious problem in this frequency band, is combatted with sufficient power margins and with coding. Link budgets, cost analysis, and weight and power calculations are also discussed. A satellite-routed FDMA system compares favorably in performance and cost with a satellite-switched TDMA system.

Berk, G.↗

Level-1 Calibration Assessment of Spire’s LEMUR-2 GNSS-R Ocean Normalized Bistatic Radar Cross Section Estimates

An assessment of the ocean surface Level-1 Normalized Bistatic Radar Cross Section products provided by the two batch 1 ‘LEMUR-2’ GNSS-R receivers of Spire, Inc. is reported. The analysis uses datasets extending from DOY 345, 2020 to DOY 329, 2021. Initial assessments indicate a highly consistent overall inter-channel response with mean NBRCS differences estimated to be at the 1.00% level over a 7-12 m/s ECMWF reference wind speed range. Efforts to validate the observation systems’ aggregate response relative to ≈3 million co-located CYGNSS measurements suggest a highly complementary behavior with an overall NBRCS correlation of 79.03% highlighting the potential utility of ‘LEMUR-2’ measurements for ocean surface wind sensing and related applications. Non-physical NBRCS dependencies on various Level-1 calibration variables, also observed with GNSS-R previous systems, are nonetheless noted and are explored in detail.

Mohammad M. Al-Khaldi↗

Elimination of space debris and objects of natural origin by laser radiation

We discuss the application of ground-based repetitively pulsed, high-frequency DF-laser systems and space-based Nd : YAG-laser systems for elimination of space debris and objects of natural origin. We have estimated the average power level of such systems ensuring destruction of space debris and similar objects. (laser applications)

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

An Optimizing Compiler for Petascale I/O on Leadership Class Architectures

In high-performance computing systems, parallel I/O architectures usually have very complex hierarchies with multiple layers that collectively constitute an I/O stack, including high-level I/O libraries such as PnetCDF and HDF5, I/O middleware such as MPI-IO, and parallel file systems such as PVFS and Lustre. Our project explored automated instrumentation and compiler support for I/O intensive applications. Our project made significant progress towards understanding the complex I/O hierarchies of high-performance storage systems (including storage caches, HDDs, and SSDs), and designing and implementing state-of-the-art compiler/runtime system technology that targets I/O intensive HPC applications that target leadership class machine. This final report summarizes the major achievements of the project and also points out promising future directions.

97 MATHEMATICS AND COMPUTING↗

Granite disposal of U.S. high-level radioactive waste.

This report evaluates the feasibility of disposing U.S. high-level radioactive waste in granite several hundred meters below the surface of the earth. The U.S. has many granite formations with positive attributes for permanent disposal. Similar crystalline formations have been extensively studied by international programs, two of which, in Sweden and Finland, are the host rocks of submitted or imminent repository license applications. This report is enabled by the advanced work of the international community to establish functional and operational requirements for disposal of a range of waste forms in granite media. In this report we develop scoping performance analyses, based on the applicable features, events, and processes (FEPs) identified by international investigators, to support generic conclusions regarding post-closure safety. Unlike the safety analyses for disposal in salt, shale/clay, or deep boreholes, the safety analysis for a mined granite repository depends largely on waste package preservation. In crystalline rock, waste packages are preserved by the high mechanical stability of the excavations, the diffusive barrier of the buffer, and favorable chemical conditions. The buffer is preserved by low groundwater fluxes, favorable chemical conditions, backfill, and the rigid confines of the host rock. An added advantage of a mined granite repository is that waste packages would be fairly easy to retrieve, should retrievability be an important objective. The results of the safety analyses performed in this study are consistent with the results of comprehensive safety assessments performed for sites in Sweden, Finland, and Canada. They indicate that a granite repository would satisfy established safety criteria and suggest that a small number of FEPs would largely control the release and transport of radionuclides. In the event the U.S. decides to pursue a potential repository in granite, a detailed evaluation of these FEPs would be needed to inform site selection and safety assessment.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Evaluation of components, subsystems, and networks for high rate, high frequency space communications

The development of new space communications technologies by NASA has included both commercial applications and space science requirements. At NASA's Lewis Research Center, methods and facilities have been developed for evaluating these new technologies in the laboratory. NASA's Systems Integration, Test and Evaluation (SITE) Space Communication System Simulator is a hardware-based laboratory simulator for evaluating space communications technologies at the component, subsystem, system, and network level, geared toward high frequency, high data rate systems. The SITE facility is well-suited for evaluation of the new technologies required for the Space Exploration Initiative (SEI) and advanced commercial systems. This paper describes the technology developments and evaluation requirements for current and planned commercial and space science programs. Also examined are the capabilities of SITE, the past, present, and planned future configurations of the SITE facility, and applications of SITE to evaluation of SEI technology.

Kerczewski, Robert J.↗

Ohmic contact on n-type Ge using Yb-germanide

Poor ohmic contact by Fermi-level pinning to valence band (E{sub V}) edge is one of the major challenges for germanium (Ge) n-type metal-oxide-semiconductor field-effect transistor (nMOSFET). Using low work-function rare-earth ytterbium (Yb), good ohmic contact on n-type Ge with alleviated Fermi-level pinning was demonstrated. Such ohmic behavior depends strongly on the germanide formation condition, where much degraded ohmic contact at 600 Degree-Sign C rapid thermal annealing is due to the lower Yb/Ge composition found by energy-dispersive x-ray spectroscopy. The ohmic behavior of Yb-germanide/n-type-Ge has high potential for future high-performance Ge nMOSFET application.

36 MATERIALS SCIENCE↗

Design and Testing of a Novel Wide Range - Segmented Gamma Scanner Incorporating Tomographic Gamma Scanning for Measuring Both Low and Intermediate Level Waste in Drums - 13470

This paper describes the design and testing of a novel automated Wide Range Segmented Gamma ray Scanning (WR-SGS) assay instrument that also incorporates Tomographic Gamma Scanning (TGS). The instrument is designed for the measurement of both Low and Intermediate Level Waste (LLW and ILW) in 200 litre drums and other waste containers covering a wide range of density. Like earlier ANTECH WR-SGS instruments, the system employs a single shielded and collimated high purity germanium (HPGe) detector to quantify the radionuclide content of the waste and like conventional SGS instruments it is suitable for the measurement of relatively homogeneous waste matrices. Also, like earlier WR-SGS systems the instrument incorporates an automated variable aperture collimator, which allows the vertical segment height to be adjusted in order to measure both high dose-rate and very low activity drums. The instrument employs both conventional discrete SGS vertical segment measurements as well as vertical segment measurement by continuous helical-scanning of the drum as it rotates. This latter method reduces measurement times for SGS measurements. In order to determine the density corrections for both low and high-density drums, a high activity Eu-152 transmission source is employed. When not in use, and in place of a conventional shutter mechanism, the shielded transmission source is moved to a shielded storage position to eliminate background radiation from the source. Due to its novel features, the WR-SGS is applicable to the measurement of both very low and very high activity waste drums as well as waste drums with a wide range of density. If located in a low background position and with the effective shielding of the strong transmission source, the instrument can be used to measure very low level or exempt waste. In order to extend the range of applicability to the measurement of heterogeneous drums, TGS measurement capability has been included in the basic WR-SGS design. This is achieved by adding horizontal motion to the waste drum rotation platform and incorporating TGS collimation into the variable aperture collimator, as well as a filter to reduce the detector signal when straight-through measurements are made using the strong transmission source. Test measurements are presented of the system operating in both SGS and TGS modes using different drum densities and source strengths. The test measurement results are compared with benchmarked MCNP Monte Carlo calculations. The Wide Range SGS - TGS instrument extends the range of both activity and density that can be measured in 200 and 340 litre drums. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Stress Testing of Chip Aluminum Polymer Capacitors

Layered aluminum polymer capacitors (APC) may be viable replacements to current chip tantalum capacitors due to their lower ESR, weight, and smaller size. So far, there is a lack of information regarding the reliability of these parts. Available technical literature is mostly related to the effect on humidity. Although APCs have a limited moisture resistance, it may be contained to the level sufficient for space applications. In this work, seven types of commercial APCs manufactured by four vendors and rated to voltages from 6.3 to 35 V were used to evaluate reliability of the parts under environmental stresses and highly accelerated life tests. Characteristics of the parts, including breakdown voltages, the level of anomalous charging currents (ACC), capacitance, DF, ESR and leakage currents were measured in the process of storage at high-temperatures (from 100 to 150 °C) and high humidity at 85 °C. Monitored highly accelerated life tests have been carried out at temperatures of 85 °C and 125 °C and voltages varying from 1.2 to 1.9 times the rated voltage. Degradation processes and reliability of APCs have been analyzed and compared to polymer tantalum capacitors.

Alexander Teverovsky↗

Leveraging Pre-Built Catalogs and Object-Level Scheduling to Eliminate I/O Bottlenecks in HPC Environments

Modern High-Performance Computing (HPC) environments face mounting challenges due to the shift from large to small file datasets, along with an increasing number of users and parallelized applications. As HPC systems rely on Parallel File Systems (PFS), such as Lustre for data processing, performance bottlenecks stemming from Object Storage Target (OST) contention have become a significant concern. Existing solutions, such as LADS with its object-level scheduling approach, fall short in large-scale HPC environments due to their inability to effectively address metadata I/O bottlenecks and the growing number of I/O processes. This study highlights the pressing need for a comprehensive solution that tackles both OST contention and metadata I/O challenges in diverse HPC workloads. To address these challenges, we propose SwiftLoad, an object-level I/O scheduling framework that leverages a metadata catalog to enhance the performance and efficiency of parallel HPC utilities. The adoption of the metadata catalog mitigates the metadata I/O bottlenecks that commonly occur in HPC utilities, a challenge that is particularly pronounced in object-level I/O scheduling. SwiftLoad addresses OST contention and the uneven distribution of I/O processes across different OSTs through mathematical modeling and incorporates a Loader Configuration Module to regulate the number of I/O processes. Evaluated with two representative utilities—data deduplication profiling and data augmentation—SwiftLoad achieved performance improvements of up to 5.63x and 11.0x, respectively, on a production supercomputer.

HPC↗

Defect Band Luminescence Intensity Reversal as Related to Application of Anti-Reflection Coating on mc-Si PV Cells: Preprint

Photoluminescence (PL) imaging is widely used to identify defective regions within mc-Si PV cells. Recent PL imaging investigations of defect band luminescence (DBL) in mc-Si have revealed a perplexing phenomenon. Namely, the reversal of the DBL intensity in various regions of mc-Si PV material upon the application of a SiNx:H anti-reflective coating (ARC). Regions with low DBL intensity beforeARC application often exhibit high DBL intensity afterwards, and the converse is also true. PL imaging alone cannot explain this effect. We have used high resolution cathodoluminescence (CL) spectroscopy and electron beam induced current (EBIC) techniques to elucidate the origin of the DBL intensity reversal. Multiple sub-bandgap energy levels were identified that change in peak position andintensity upon the application of the ARC. Using this data, in addition to EBIC contrast information, we provide an explanation for the DBL intensity reversal based on the interaction of the detected energy levels with the SiNx:H ARC application. Multiple investigations have suggested that this is a global problem for mc-Si PV cells. Our results have the potential to provide mc-Si PV producers apathway to increased efficiencies through defect mitigation strategies.

14 SOLAR ENERGY↗

Object Proxy Patterns for Accelerating Distributed Applications

Workflow and serverless frameworks have empowered new approaches to distributed application design by abstracting compute resources. However, their typically limited or one-size-fits-all support for advanced data flow patterns leaves optimization to the application programmer—optimization that becomes more difficult as data become larger. The transparent object proxy, which provides wide-area references that can resolve to data regardless of location, has been demonstrated as an effective low-level building block in such situations. Here we propose three high-level proxy-based programming patterns—distributed futures, streaming, and ownership—that make the power of the proxy pattern usable for more complex and dynamic distributed program structures. We motivate these patterns via careful review of application requirements and describe implementations of each pattern. As a result, we evaluate our implementations through a suite of benchmarks and by applying them in three meaningful scientific applications, in which we demonstrate substantial improvements in runtime, throughput, and memory usage.

Distributed Computing↗