Search NASA⌕ Search

SEARCH · Search NASA

Results for “HLS”

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 73 records · Page 4

MLIR loop optimizations for High-Level Synthesis: a case study

High-Level Synthesis (HLS) tools simplify the design of hardware accelerators by automatically generating Verilog/VHDL code starting from a general purpose software programming language. They include a wide range of optimization techniques in the process, most of them performed on a low-level intermediate representation (IR) of the code. Introducing optimizations on a higher level of abstraction could significantly contribute to the automated design process results; for example, polyhedral techniques for the manipulation of loops could have a significant impact on the generated accelerators when applied on a specialized IR. We use loop pipelining as a case study to explore the introduction of compiler-based transformations on top of an existing HLS process. We leverage the Multi-Level Intermediate Representation (MLIR) framework and an external scheduler to implement the required transformations, and couple them with existing HLS tools to evaluate the improvements that loop pipelining brings to the performance of generated accelerators. The proposed approach can be integrated with other high-level transformations on the MLIR representation, combining different techniques to obtain pre-optimized inputs for HLS that do not have to rely on a specific backend tool.

Curzel, Serena↗

Evaluation of the FIR Example using Xilinx Vivado High-Level Synthesis Compiler

Compared to central processing units (CPUs) and graphics processing units (GPUs), field programmable gate arrays (FPGAs) have major advantages in reconfigurability and performance achieved per watt. This development flow has been augmented with high-level synthesis (HLS) flow that can convert programs written in a high-level programming language to Hardware Description Language (HDL). Using high-level programming languages such as C, C++, and OpenCL for FPGA-based development could allow software developers, who have little FPGA knowledge, to take advantage of the FPGA-based application acceleration. This improves developer productivity and makes the FPGA-based acceleration accessible to hardware and software developers. Xilinx Vivado HLS compiler is a high-level synthesis tool that enables C, C++ and System C specification to be directly targeted into Xilinx FPGAs without the need to create RTL manually. The white paper [1] published recently by Xilinx uses a finite impulse response (FIR) example to demonstrate the variable-precision features in the Vivado HLS compiler and the resource and power benefits of converting floating point to fixed point for a design. To get a better understanding of variable-precision features in terms of resource usage and performance, this report presents the experimental results of evaluating the FIR example using Vivado HLS 2017.1 and a Kintex Ultrascale FPGA. In addition, we evaluated the half-precision floating-point data type against the double-precision and single-precision data type and present the detailed results.

97 MATHEMATICS AND COMPUTING↗

Even Higher-Level Synthesis: An Exploration of AI Hardware Accelerators using HLS4ML

With the rise of artificial intelligence, the popularization of deep learning, and a constantly evolving industry, the demand for flexible and efficient tools has never been greater. As algorithms grow more complex, their runtime and energy consumption increase exponentially. Customized hardware accelerators, long used for specific mathematical operations, remain essential for managing modern applications' computational and power demands. Hardware accelerators can speed up complex computations by orders of magnitude, but their manual design and verification processes are often challenging and time-consuming. High-Level Synthesis (HLS) provides a solution by transforming high-level algorithm descriptions, typically written in C++ or SystemC, into synthesizable RTL suitable for hardware implementation. This approach reduces development time for RTL engineers while offering flexibility beyond what traditional handwritten RTL can provide. We extended this capability to the machine-learning domain with the open-source framework hls4ml, which allows neural networks trained in Python frameworks like Tensorflow or PyTorch to be synthesized into efficient hardware representations for the traditional FPGA and ASIC flows. This breakthrough addresses the growing need for reduced design turnaround and easy verification of ML hardware accelerators with low latency and power efficiency constraints. During this tutorial, we will demonstrate how Python complements HLS by simplifying the ML design process, bridging the gap between software and hardware development. Attendees will explore how we translate neural networks modeled in Python into fixed-point C++ models suitable for HLS workflows. We will dive into strategies like Value-Range Analysis and Quantization-Aware Training, which optimize these designs for deployment and evaluate their accuracy, power consumption, and energy efficiency. To exemplify these concepts, experts from Fermilab will share their experiences applying this technology to high-energy physics experiments, where real-time, low-latency processing is critical. Over the years, Fermilab engineers have demonstrated how deep neural networks, optimized for hardware using hls4ml, can meet the stringent requirements of trigger systems at the CERN Large Hadron Collider. These systems rely on rapid decision-making to process immense data volumes while retaining only the most relevant events for further analysis. The application of hls4ml has also been extended to innovative technologies like smart pixel arrays. These smart pixels integrate ML inference capabilities directly into sensor devices, enabling localized data processing at the pixel level. This approach drastically reduces the need to transmit raw data to external processing units, significantly decreasing power consumption and latency. By embedding neural networks within the pixel architecture, the smart pixels can identify and prioritize relevant data in real time, providing a highly efficient solution for edge computing in scenarios such as particle detectors and imaging systems. Fermilab's work highlights the potential of hardware-accelerated ML in scenarios where both speed and power efficiency are mission-critical. Through this tutorial, attendees will gain valuable insights into the challenges and solutions of deploying ML in hardware. Understanding how HLS and hls4ml streamline the development of neural network-based hardware accelerators is fundamental for the industry's future. Participants will learn how these technologies are shaping the future of AI and scientific computing.

Di Guglielmo, Giuseppe [Fermilab]↗

The effect of purified compared with nonpurified diet on bone changes induced by hindlimb suspension of female rats

The purpose of this study was to compare the bone changes induced by unloading in rats fed different diets, because space flight studies use a semipurified diet, whereas space flight simulation studies typically use nonpurified diets. Female Sprague-Dawley rats were fed a purified American Institute of Nutrition (AIN) 93G diet or a standard nonpurified diet and kept ambulatory or subjected to unloading by hindlimb suspension (HLS) for 38 days. Bone mineral content (BMC), mechanical strength, and factors related to the diet that affect bone (i.e., urinary calcium excretion, estradiol, and corticosterone) were measured. Average food intakes (grams per day) differed for diets, but caloric intake (kilocalories per day) and the final body masses of treatment groups were similar. The HLS-induced decrease in femoral BMC was not statistically different for rats fed a nonpurified diet (-8.6%) compared with a purified AIN-93G diet (-11.4%). The HLS-induced decrease in femoral mechanical strength was not statistically different for rats fed a nonpurified diet (-24%) compared with a purified AIN-93G diet (-31%). However, bone lengths were decreased (P < 0.05) in rats fed a nonpurified diet compared with a purified diet. Plasma estradiol levels were lower (P < 0.05) in the HLS/AIN-93G group but similar in the HLS and ambulatory rats fed a nonpurified diet. Plasma estradiol was related to femoral BMC (r = 0.85, P < 0.01). Urinary calcium excretion was higher (P < 0.05) in rats fed a nonpurified diet than those fed a purified AIN-93G diet, which is consistent with the higher level of calcium in the nonpurified diet. Urinary corticosterone levels were higher (P < 0.05) in rats fed a nonpurified diet than rats fed the AIN-93G diet. Although the osteopenia induced by unloading was similar in both diet groups, there were differences in longitudinal bone growth, calcium excretion, plasma estradiol levels, and urinary corticosterone levels. Results indicate that the type of standard diet used is an important factor to consider when measuring bone end points.

Bone and Bones/pathology↗

Harmonized Landsat/Sentinel-2 Products for Land Monitoring

The Harmonized Landsat-8 and Sentinel-2 (HLS) project is a NASA initiative aiming to produce a seamless, harmonized surface reflectance record from the Operational Land Imager (OLI) and Multi-Spectral Instrument (MSI) aboard Landsat-8 and Sentinel-2 remote sensing satellites, respectively. The HLS products are based on a set of algorithms to obtain seamless products from both sensors (OLI and MSI): atmospheric correction, cloud and cloud-shadow masking, geographic co-registration and common gridding, bidirectional reflectance distribution function normalization and bandpass adjustment. As of version 1.3, the HLS v1.3 data set covers 9.12 million km2 and spans from first Landsat-8 data (2013) to present. HLS products provide near-daily surface reflectance information with a common geometric framework, and are suitable for a variety of agricultural and vegetation monitoring tasks, including analysis of crop type, condition, and phenology.

Masek, Jeffrey↗

NASA's Artemis Human Landing Systems

The Human Landing System (HLS) is the mode of transportation that will take astronauts to the lunar surface in NASA's Artemis lunar exploration program. On early missions, the astronauts will live inside the pressurized crew cabin portion of the lander for up to a week. The HLS program, based at Marshall Space Flight Center in Huntsville, Alabama, is working closely with commercial partners to build innovative and technically advanced lunar landers, leveraging decades of human spaceflight experience and the speed of the commercial sector. In addition to the NextSTEP-2 Broad Agency Announcement Appendix H Option A contract that includes one uncrewed and one crewed demonstration mission to the lunar surface, the HLS program is taking steps to establish a regular cadence of crewed missions to the surface of the Moon. This paper will discuss the HLS program’s latest development activities.

Lisa Watson-Morgan↗

NASA’s Artemis Human Landing Systems

The Human Landing System (HLS) is the mode of transportation that will take astronauts to the lunar surface in NASA's Artemis lunar exploration program. On early missions, the astronauts will live inside the pressurized crew cabin portion of the lander for up to a week. The HLS program, based at NASA’s Marshall Space Flight Center in Huntsville, Alabama, is working closely with commercial partners to build innovative and technically advanced lunar landers, leveraging decades of human spaceflight experience and the speed of the commercial sector. In addition to the NextSTEP-2 Broad Agency Announcement Appendix H Option A contract that includes one uncrewed and one crewed demonstration mission to the lunar surface, the HLS program is taking steps to establish a regular cadence of crewed missions to the surface of the Moon. This paper will discuss the HLS program’s latest development activities.

Lisa Watson-Morgan↗

Artemis: An Overview of NASA’s Activities to Return Humans to the Moon

NASA is well underway in its plans to return humans to the Moon and build long-term infrastructure both in orbit and on the surface. Following the Artemis I and Artemis II flight tests, NASA will begin crewed landings and regular missions to the Moon. Artemis will return value to the American public and the global community, enable groundbreaking scientific discovery, and prepare NASA and its partners for exploration of the solar system. Work on the initial Human Landing System (HLS), procurement activities for future HLS, and work on the initial modules of the lunar-orbiting Gateway are underway. International partnerships are playing a critical role in the planning, development, and execution of these missions. As of December 2021, three international partners have entered agreements to contribute to the Gateway and 14 nations have signed the Artemis accords. A new spacesuit acquisition strategy will stimulate economic growth and meet the growing needs of NASA and emerging customers. Requirements development is underway for an unpressurized lunar surface rover and plans for a pressurized rover, foundation surface habitat, and the Artemis Base Camp are in work. This paper will provide an update on the HLS demonstration missions and the procurement status for HLS services; a comprehensive look at the Gateway’s components and related statuses; a summary of planned surface systems; and an outline of the strategies NASA will use to successfully implement these plans, preparing for humanity's next giant leap: Mars.

Artemis↗

An Orbit Determination Comparison Study and Demonstration for Rendezvous and Docking in a Near Rectilinear Halo Orbit from the Lunar Surface

For the upcoming NASA Artemis III mission and those that follow, both the Human Landing System (HLS) and Orion programs are invested in understanding the impacts of ground tracking performance in supporting rendezvous and docking in a Near Rectilinear Halo Orbit (NRHO). Several critical questions must be answered to ensure mission success and crew safety and an assortment of analysis tools are being incorporated to address them. Two of these tools, LINCOV and MONTE, are currently providing program decision making results through HLS Insight, HLS NASA-collaborations, and Orion/Gateway cross-program analysis. To ensure consistency in the orbit determination performance, a comparison trade-study is performed using a low-lunar orbit to NRHO rendezvous scenario anticipated for the upcoming Artemis missions. An overview of the two analysis tools is provided along with a detailed step-by-step evaluation of the core capabilities and models related to the orbit determination process. This incremental comparison effort reveals both tools produce consistent solutions for the criteria investigated. Given the confidence in the orbit determination process and solutions generated, these results are then applied to demonstrate an integrated, closed-loop system performance where the HLS lander ascends from the lunar surface and successfully inserts into the NRHO relative to the Orion spacecraft in preparation for the final rendezvous and docking phase.

Linear Covariance Analysis↗

Analysis of the Approach Direction of the Human Landing System to a South Pole Landing Site Subject to Lighting and Communication Constraints

The NRHO used as the staging orbit for crewed missions to the lunar surface by the HLS remains nearly perpendicular to the Earth-Moon line of sight, however, due to the libration of the Moon the location of the South Pole is ever shifting beneath it. DTE communication as well as ground illumination conditions near the South Pole coincide with lunar libration, with approximately 2-week cycles of DTE communication and solar availability. Additionally, the constant shifting of the South Pole region with respect to the NRHO leads to variations in the nominal HLS approach path to the surface in order to maintain near in-plane alignment of the NRHO, LLO, and approach phase. Making adjustments to the nominal approach path to achieve desired communication and lighting conditions quickly raises in-space ∆V. This analysis compares the ∆V impact of several trajectory strategies that could be used to alter the HLS approach path.

HLS↗

Horn Location Sensors for LBNF

The Horn Location Sensor (HLS) system—named for the magnetic focusing horns it monitors, along with other critical beamline elements – is a high-precision alignment system developed for the Long-Baseline Neutrino Facility (LBNF) to support the Deep Underground Neutrino Experiment (DUNE). With minimal maintenance, the HLS can operate reliably in environments with high radiation, and it ensures that key components such as the protective baffle, focusing horns, and beam position monitors are aligned correctly – each essential for maintaining one of the world’s most intense muon-neutrino beams. A high-precision hydrostatic level sensor, a linear variable differential transformer, and INVAR rods are all used in the system to monitor vertical motion and tilt with sub-millimeter accuracy. A novel Sweep Tracker interferometer enhances calibration fidelity by correcting for non-linearities in laser wavelength and scan rate in real time. A critical part of DUNE’s precision alignment and flux prediction requirements, the HLS system initially supports beam power of up to 1.2 MW and can be upgraded to 2.4 MW.

Frequency scanning interferometry↗

Strong, ductile, and hierarchical hetero-lamellar-structured alloys through microstructural inheritance and refinement

The strength−ductility trade-off exists ubiquitously, especially in brittle intermetallic-containing multiple principal element alloys (MPEAs), where the intermetallic phases often induce premature failure leading to severe ductility reduction. Hierarchical heterogeneities represent a promising microstructural solution to achieve simultaneous strength−ductility enhancement. However, it remains fundamentally challenging to tailor hierarchical heterostructures using conventional methods, which often rely on costly and time-consuming processing. Here, we report a multiscale microstructural inheritance and refinement strategy to process “structural hierarchy precursors” in as-cast heterogeneous Al 0.7 CoCrFeNi MPEAs, which lead directly to a hierarchical hetero-lamellar structure (HLS) after simple rolling and annealing. Interestingly, it takes only 10 min of annealing time, two orders of magnitude less than that required to render the state-of-the-art properties during conventional processing of Al 0.7 CoCrFeNi, for us to achieve record-high strength−ductility combinations via the hierarchical HLS design that sequentially stimulates multiple unusual deformation and reinforcement mechanisms. In particular, the HLS-enabled high hetero-deformation-induced (HDI) internal stress triggers profuse <111>-type dislocations on over five independent slip systems in the supposedly brittle intermetallic phase and activates extensive stacking faults (SFs) and nanotwinning in the adjoining soft phase with a rather high SF energy. These unexpected, dynamically reinforcing hetero-deformation mechanisms across multiple length scales facilitate high sustained HDI strain hardening, along with a salient microcrack-mediated extrinsic ductilization effect, suggesting that the proposed microstructural inheritance and refinement strategy provides an efficient, fast, and low-cost approach to overcome the strength−ductility trade-off in a broad range of structural materials.

Science & Technology - Other Topics↗

Effective Lagrangian for low-scale technicolor

We present an effective Lagrangian for low-scale technicolor. It describes the interactions at energies < or approx. M{sub {rho}{sub T}} of the lowest-lying bound states of the lightest technifermion doublet--the spin-one {rho}{sub T}, {omega}{sub T}, a{sub T}, f{sub T}, and the corresponding technipions {pi}{sub T}. This Lagrangian is intended to put on firmer ground the technicolor straw man phenomenology used for collider searches of low-scale technicolor. The technivectors are described using the hidden local symmetry (HLS) formalism of Bando, et al. The Lagrangian is based on SU(2) x U(1) x U(2){sub L} x U(2){sub R}, where SU(2) x U(1) is the electroweak gauge group and U(2){sub L} x U(2){sub R} is the HLS gauge group. Special attention is paid to the higher-derivative standard HLS and Wess-Zumino-Witten interactions needed to describe radiative and other decays of a{sub T} and {rho}{sub T}/{omega}{sub T}, respectively.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Extending High-Level Synthesis with AI/ML Methods

Artificial Intelligence (AI) and Machine Learning (ML) methods provide significant opportunities of improving quality of results when performing high-level synthesis (HLS). For example, they can be used to model and predict metrics of the final design (e.g., area, considering aspects such as interconnect overhead for different device technologies), facilitating exploration when searching for the best design trade-offs. They can also enable identifying hidden correlations across the various phases of the synthesis and the various optimizations performed, identifying the most effective pipelines. Finally, in more general terms, bio-inspired heuristic algorithms can improve the design space exploration for the synthesis process in terms of time and quality of the result. This paper discusses opportunities and challenges to augment HLS with AI/ML using as example flow the SODA Synthesizer, an open-source hardware generation toolchain which includes SODA-OPT, a hardware/software partitioning and pre-optimization tool developed with the MLIR framework, and PandA-Bambu, a state-of-the art HLS tool. SODA interfaces with OpenROAD to provide a complete end-to-end toolchain.

artificial intelligence↗

Ground Movement in SSRL Ring

Users of the Stanford Synchrotron Radiation Lightsource (SSRL) are being affected by diurnal motion of the synchrotron's storage ring, which undergoes structural changes due to outdoor temperature fluctuations. In order to minimize the effects of diurnal temperature fluctuations, especially on the vertical motion of the ring floor, scientists at SSRL tried three approaches: painting the storage ring white, covering the asphalt in the middle of the ring with highly reflective Mylar and installing Mylar on a portion of the ring roof and walls. Vertical motion in the storage ring is measured by a Hydrostatic Leveling System (HLS), which calculates the relative height of water in a pipe that extends around the ring. The 24-hr amplitude of the floor motion was determined using spectral analysis of HLS data, and the ratio of this amplitude before and after each experiment was used to quantitatively determine the efficacy of each approach. The results of this analysis showed that the Mylar did not have any significant effect on floor motion, although the whitewash project did yield a reduction in overall HLS variation of 15 percent. However, further analysis showed that the reduction can largely be attributed to a few local changes rather than an overall reduction in floor motion around the ring. Future work will consist of identifying and selectively insulating these local regions in order to find the driving force behind diurnal floor motion in the storage ring.

43 PARTICLE ACCELERATORS↗

Verifiable Security Templates for Hardware

High-level synthesis (HLS) research generally focuses on transferring "software engineering virtues" (e.g., modularity, abstraction, extensibility, etc.) to hardware development with the ultimate goal of making hardware development as agile as software development. And recent HLS research has focused on transferring ideas and techniques from high assurance software formal methods to hardware development. Just as it has introduced software engineering virtues, we believe HLS can also become a vector for adapting software formal methods to the challenge of high assurance security in hardware. This paper introduces the Device Calculus and its mechanization in the Agda proof checking system. The Device Calculus is a starting point for exploring the formal methods and security of high-level synthesis flows. We illustrate the Device Calculus with a number of examples of formally verifiable security templates-i.e., functions in the Device Calculus that express common security structures at a high-level of abstraction.

Harrison, Bill↗

Stimulation of Myofibrillar Protein Synthesis in Hindlimb Suspended Rats by Resistance Exercise and Growth Hormone

The objective of this study was to determine the ability of a single bout of resistance exercise alone or in combination with recombinant human growth hormone (rhGH) to stimulate myofibrillar protein synthesis (Ks) in hindlimb suspended (HLS) adult female rats. Plantar flexor muscles were stimulated with resistance exercise, consisting of 10 repetitions of ladder climbing on a 1 m grid (85 deg.), carrying an additional 50% of their body weight attached to their tails. Saline or rhGH (1 mg/kg) was administered 30' prior to exercise, and Ks was determined with a constant infusion of H-3-Leucine at 15', 60', 180', and 360' following exercise. Three days of HLS depressed Ks is approx. equal to 65% and 30-40% in the soleus and gastrocnemius muscles, respectively (p is less than or equal to 0.05). Exercise increased soleus Ks in saline-treated rats 149% 60' following exercise (p less than or equal to 0.05), decaying to that of non-exercised animals during the next 5 hours. Relative to suspended, non-exercised rats rhGH + exercise increased soleus Ks 84%, 108%, and 72% at 15', 60' and 360' following exercise (p is less than or equal to 0.05). Gastrocnemius Ks was not significantly increased by exercise or the combination of rhGH and exercise up to 360' post-exercise. Results from this study indicate that resistance exercise stimulated Ks 60' post-exercise in the soleus of HLS rats, with no apparent effect of rhGH to enhance or prolong exercise-induced stimulation. Results suggests that exercise frequency may be important to maintenance of the slow-twitch soleus during non-weightbearing, but that the ability of resistance exercise to maintain myofibrillar protein content in the gastrocnemius of hindlimb suspended rats cannot be explained by acute stimulation of synthesis.

Linderman, Jon K.↗

Bone Blood Flow During Simulated Microgravity: Physiological and Molecular Mechanisms

Blood flow to bone has been shown to affect bone mass and presumably bone strength. Preliminary data indicate that blood flow to the rat femur decreases after 14 days of simulated microgravity, using hindlimb suspension (HLS). If adult rats subjected to HLS are given dobutamine, a synthetic catecholamine which can cause peripheral vasodilation and increased blood flow, the loss of cortical bone area usually observed is prevented. Further, mechanisms exist at the molecular level to link changes in bone blood flow to changes in bone cell activity, particularly for vasoactive agents like nitric oxide (NO). The decreases in fluid shear stress created by fluid flow associated with the shifts of plasma volume during microgravity may result in alterations in expression of vasoactive agents such as NO, producing important functional effects on bone cells. The primary aim of this project is to characterize changes in 1) bone blood flow, 2) indices of bone mass, geometry, and strength, and 3) changes in gene expression for modulators of nitric oxide activity (e.g., nitric oxide synthase) and other candidate genes involved in signal transduction of mechanical loading after 3, 7, 14, 21, and 28 days of HLS in the adult rat. Using a rat of at least 5 months of age avoids inadvertently studying effects of simulated microgravity on growing, rather than adult, bone. Utilizing the results of these studies, we will then define how altered blood flow contributes to changes in bone with simulated microgravity by administering a vasodilatory agent (which increases blood flow to tissues) during hindlimb suspension. In all studies, responses in the unloaded hindlimb bones (tibial shaft, femoral neck) will be compared with those in the weightbearing humeral shaft and the non-weightbearing calvarium (skull) from the same animal. Bone volumetric mineral density and geometry will be quantified by peripheral quantitative CT; structural and material properties of the long bones will be determined by 3-point bending (tibia, humerus) or compression (femoral neck) testing to failure. A unique aspect of these studies will be defining the time course of changes in gene expression in bone cell populations with unloading, accomplished with Northern blots, in situ hybridization, and immunohistochemistry. These studies have high relevance for concurrent protocols being proposed by investigators on NSBRI Cardiovascular and Muscle teams, with blood flow data available on a number of tissues other than bone. Further, dobutamine and other Beta-agonists have been tested as countermeasures for altered muscle and cardiovascular function. Results of the intervention tested in our studies have potential relevance for a number of systemic changes seen with prolonged spaceflight.

Bloomfield, Susan A.↗