Search NASA⌕ Search

SEARCH · Search NASA

Results for “Network Automation”

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 1,405 records · Page 78

Third Annual Workshop on Space Operations Automation and Robotics (SOAR 1989)

Papers presented at the Third Annual Workshop on Space Operations Automation and Robotics (SOAR '89), hosted by the NASA Lyndon 8. Johnson Space Center at Houston, Texas, on July 25-27, 1989, are documented herein. During the three days, approximately 100 technical papers wer3 presented by experts from NASA, the USAF, universities, and technical companies. Also held were panel discussions on Air Force/NASA AI Overview and Expert System Verification and Validation. Tutorial sessions included Neural Networks; Theory and Application of Back Propagation; Verification and Validation of Expert Systems/ Evaluation of Expert System Tools; and Technical Environment for Modular Architectures for Robotics in Space; and are not documented herein. Technical topics addressed included intelligent systems, robotics, human factors, and environment.

Knowledge representation↗

Railroad Valley Radiometric Calibration Test Site (RadCaTS) as Part of a Global Radiometric Calibration Network (RadCalNet)

The Radiometric Calibration Network (RadCalNet) is a coordinated multinational effort to provide in situ data that are suitable for the radiometric calibration and validation of Earth observation sensors that operate in the visible to shortwave infrared solar reflective spectral region (400 nm to 1000 nm). The main goals of RadCalNet are to provide top-of-atmosphere reflectance data to the scientific community, standardize data collection protocols for automated test sites, and to document the SI-traceable uncertainty budgets for each automated test site, of which there are currently four. The data available from RadCalNet are suitable for the calibration and validation of spaceborne imaging spectrometers. The work presented here provides a description of RadCalNet as well as a sample of the current results from the Radiometric Calibration Test Site (RadCaTS), which is located at Railroad Valley, Nevada, USA. Selected sensors for comparison include Terra and Aqua MODIS, SNPP and NOAA-20 VIIRS, and Sentinel-3A and -3B OLCI.

RadCalNet↗

Distributed Visual Sensing and Fusion for Advanced Air Mobility

Surveillance solutions for Advanced and Urban Air Mobility frameworks are a key factor to enable safe operations of highly automated aircraft in the civil airspace. To design solutions suitable for all types of aircraft, non-cooperative sensors can be used, though many challenges arise when the small dimensions of the vehicles and their proximity to the ground during low-altitude missions are considered. A distributed sensing concept can be efficiently applied to address these challenges by exploiting multiple sensors within a surveillance network. This paper proposes a strategy to fuse the information collected by three ground-fixed cameras within a network of multiple distributed sensors and is tested with during experimental flight tests. The solution exploits standalone tracking estimates of each camera within a fusion center that performs triangulation and three-dimensional tracking. This approach is tested in a scenario involving two small UAVs flying at low altitude. The paper deals with the challenges of associating the two objects from independent and unrelated tracks to achieve robust triangulation, which produces meter-level mean errors with respect to GNSS-based ground truth.

Federica Vitiello↗

Data Access Services that Make Remote Sensing Data Easier to Use

This slide presentation reviews some of the processes that NASA uses to make the remote sensing data easy to use over the World Wide Web. This work involves much research into data formats, geolocation structures and quality indicators, often to be followed by coding a preprocessing program. Only then are the data usable within the analysis tool of choice. The Goddard Earth Sciences Data and Information Services Center is deploying a variety of data access services that are designed to dramatically shorten the time consumed in the data preparation step. On-the-fly conversion to the standard network Common Data Form (netCDF) format with Climate-Forecast (CF) conventions imposes a standard coordinate system framework that makes data instantly readable through several tools, such as the Integrated Data Viewer, Gridded Analysis and Display System, Panoply and Ferret. A similar benefit is achieved by serving data through the Open Source Project for a Network Data Access Protocol (OPeNDAP), which also provides subsetting. The Data Quality Screening Service goes a step further in filtering out data points based on quality control flags, based on science team recommendations or user-specified criteria. Further still is the Giovanni online analysis system which goes beyond handling formatting and quality to provide visualization and basic statistics of the data. This general approach of automating the preparation steps has the important added benefit of enabling use of the data by non-human users (i.e., computer programs), which often make sub-optimal use of the available data due to the need to hard-code data preparation on the client side.

Lynnes, Christopher↗

Computational Workflow for Accelerated Molecular Design Using Quantum Chemical Simulations and Deep Learning Models

Efficient methods for searching the chemical space of molecular compounds are needed to automate and accelerate the design of new functional molecules such as pharmaceuticals. Given the high cost in both resources and time for experimental efforts, computational approaches play a key role in guiding the selection of promising molecules for further investigation. Here, we construct a workflow to accelerate design by combining approximate quantum chemical methods [i.e. density-functional tight-binding (DFTB)], a graph convolutional neural network (GCNN) surrogate model for chemical property prediction, and a masked language model (MLM) for molecule generation. Property data from the DFTB calculations are used to train the surrogate model; the surrogate model is used to score candidates generated by the MLM. The surrogate reduces computation time by orders of magnitude compared to the DFTB calculations, enabling an increased search of chemical space. Furthermore, the MLM generates a diverse set of chemical modifications based on pre-training from a large compound library. We utilize the workflow to search for near-infrared photoactive molecules by minimizing the predicted HOMO-LUMO gap as the target property. Our results show that the workflow can generate optimized molecules outside of the original training set, which suggests that iterations of the workflow could be useful for searching vast chemical spaces in a wide range of design problems.

Blanchard, Andrew↗

Direct versus Facility Centric Load Control for Automated Demand Response

Direct load control (DLC) refers to the scenario where third party entities outside the home or facility are responsible for deciding how and when specific customer loads will be controlled in response to Demand Response (DR) events on the electric grid. Examples of third parties responsible for performing DLC may be Utilities, Independent System Operators (ISO), Aggregators, or third party control companies. DLC can be contrasted with facility centric load control (FCLC) where the decisions for how loads are controlled are made entirely within the facility or enterprise control systems. In FCLC the facility owner has more freedom of choice in how to respond to DR events on the grid. Both approaches are in use today in automation of DR and both will continue to be used in future market segments including industrial, commercial and residential facilities. This paper will present a framework which can be used to differentiate between DLC and FCLC based upon where decisions are made on how specific loads are controlled in response to DR events. This differentiation is then used to compare and contrast the differences between DLC and FCLC to identify the impact each has on:(1)Utility/ISO and third party systems for managing demand response, (2)Facility systems for implementing load control, (3)Communications networks for interacting with the facility and (4)Facility operators and managers. Finally a survey of some of the existing DR related specifications and communications standards is given and their applicability to DLC or FCLC. In general FCLC adds more cost and responsibilities to the facilities whereas DLC represents higher costs and complexity for the Utility/ISO. This difference is primarily due to where the DR Logic is implemented and the consequences that creates. DLC may be more certain than FCLC because it is more predictable - however as more loads have the capability to respond to DR signals, people may prefer to have their own control of end-use loads and FCLC systems. Research is needed to understand the predictability of FCLC which is related to the perceived value of the DR from the facility manager or home owner's perspective.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Graph dynamical networks for unsupervised learning of atomic scale dynamics in materials

Understanding the dynamical processes that govern the performance of functional materials is essential for the design of next generation materials to tackle global energy and environmental challenges. Many of these processes involve the dynamics of individual atoms or small molecules in condensed phases, e.g. lithium ions in electrolytes, water molecules in membranes, molten atoms at interfaces, etc., which are difficult to understand due to the complexity of local environments. In this work, we develop graph dynamical networks, an unsupervised learning approach for understanding atomic scale dynamics in arbitrary phases and environments from molecular dynamics simulations. We show that important dynamical information, which would be difficult to obtain otherwise, can be learned for various multi-component amorphous material systems. With the large amounts of molecular dynamics data generated every day in nearly every aspect of materials design, this approach provides a broadly applicable, automated tool to understand atomic scale dynamics in material systems.

36 MATERIALS SCIENCE↗

Correction and calibration of atmospheric impact observations in GOES GLM data

The Earth's atmosphere is impacted daily by both meteoroids and artificial objects. Calibrated observations of the emitted light at sufficiently high sampling rates can enable or improve the estimation of impactor attributes such as size, cohesion, trajectory, and composition, but are difficult to obtain owing to the unpredictability, brevity, and high dynamic (brightness) range of impacts. Ground-based camera systems have successfully monitored small regions of the atmosphere at video frame rates and with limited radiometric capabilities, but most impacts occur over the 70% of the Earth's surface covered by water and are therefore missed by these networks. The Geostationary Lightning Mapper (GLM) instruments aboard Geostationary Operational Environmental Satellites 16 and 17 provide near-hemispherical coverage at 500 frames per second. These data have been shown to contain the signatures of many independently confirmed impacts, often from both viewing angles simultaneously, and constitute an observational resource that is currently unparalleled in the public domain. NASA's Asteroid Threat Assessment Project has implemented an automated impact detection pipeline that processes data from GLM daily. Given a detected impact, the GLM data contain a wealth of information for use in quantitative follow-up analyses. However, impact events differ from lightning in ways that violate key assumptions built into GLM's design. The result is that GLM's onboard processing introduces errors into pixel observations of impact events and the calibrated energies near the periphery of the detector may be substantially overestimated. We present methods for mitigating these and other issues to produce a data product more suitable for impact analyses than the existing GLM lightning product.

58 GEOSCIENCES↗

Performance Comparison of Machine Learning Models for Ultrasonic Nondestructive Evaluation of Alkali-Silica Reaction in Concrete

Alkali-silica reaction (ASR) causes concrete degradation, leading to cracking, rebar corrosion, and reduced structural integrity, which raises safety concerns. Ultrasonic nondestructive evaluation (NDE) effectively assesses concrete properties and monitors ASR progression. However, its deployment and analysis require specialized expertise and subjective interpretation. As computational power increases, artificial intelligence (AI) and machine learning (ML) algorithms are increasingly being used to automate NDE data analysis across various industries for AI-assisted automation. Regulatory agencies are adapting to this technological shift, prompting a need to evaluate current ML technologies’ capabilities and limitations in assessing concrete material properties and damage. This report presents a comparative analysis of four ML regression models for predicting concrete material damage induced by ASR expansion using long-term ultrasonic data monitoring. The models investigated include linear regression (LR), support vector regression (SVR), shallow neural networks (NN), and deep neural networks (DNN). LR, SVR, and shallow NN models use features extracted from ultrasonic signals, whereas the DNN model processes time-domain ultrasonic signals and frequency spectra directly. The study systematically compared the models’ performance from various perspectives, including model input, prediction performance, and generalization ability. The findings indicate significant variability in model performance, with some ML algorithms achieving very high or very low prediction accuracy depending on the preprocessing and feature engineering (extraction and selection) applied. Key insights include the observation that shallow ML models (LR, SVR, and shallow NNs) require meticulous preprocessing and feature extraction to achieve high accuracy. In contrast, the DNN model, although it bypasses the need for feature engineering, necessitates extensive preprocessing to mitigate noise and computational demands. The SVR model emerged as the top performer among the shallow models, and the DNN model exhibited superior performance on specific datasets but struggled with generalization across specimens from different batches. Additionally, the SVR model is sensitive to temperature variations, whereas the DNN model is robust in this regard. Using recurrent neural networks is recommended for future ASR expansion prediction studies. Recurrent neural networks’ inherent ability to capture temporal dependencies and long-term patterns makes them well suited for analyzing sequential ultrasonic monitoring data. Overall, the results and conclusions of this study could provide insights into the capabilities and effectiveness of ML when applied to ultrasonic NDE data and help identify best practices for using ML for ultrasonic NDE of concrete material properties.

36 MATERIALS SCIENCE↗

Mars Reconnaissance Orbiter, Ground Data System, Receivables and Deliverables (REC/DELs)

This paper presents one JPL element manager's approach to describe a complex Ground Data System (GDS) with its receivables and deliverables (REC/DEL). The Mars Reconnaissance Orbiter (MRO) Ground Data System is the integrated set of ground software, hardware, facilities and networks that support mission operation. REC/DEL is a powerful tool for specifying hierarchy of commitments among systems and teams. Receivable of a system is a deliverable of another system. Focusing on tangible products enables the manager to objectively measure progress in a schedule. Jet Propulsion Laboratory mandates the use of REC/DEL for flight projects. Tutorial and training is provided for managers to create integrated REC/DEL database using automated systems. Project schedules are based on REC/DELs. This paper is not focusing on the mechanics of REC/DEL database creation, but it provides a guideline how one systematically creates categories of deliverables and receivables for ground data system components.

guidelines↗

VESGEN Software for Mapping and Quantification of Vascular Regulators

VESsel GENeration (VESGEN) Analysis is an automated software that maps and quantifies effects of vascular regulators on vascular morphology by analyzing important vessel parameters. Quantification parameters include vessel diameter, length, branch points, density, and fractal dimension. For vascular trees, measurements are reported as dependent functions of vessel branching generation. VESGEN maps and quantifies vascular morphological events according to fractal-based vascular branching generation. It also relies on careful imaging of branching and networked vascular form. It was developed as a plug-in for ImageJ (National Institutes of Health, USA). VESGEN uses image-processing concepts of 8-neighbor pixel connectivity, skeleton, and distance map to analyze 2D, black-and-white (binary) images of vascular trees, networks, and tree-network composites. VESGEN maps typically 5 to 12 (or more) generations of vascular branching, starting from a single parent vessel. These generations are tracked and measured for critical vascular parameters that include vessel diameter, length, density and number, and tortuosity per branching generation. The effects of vascular therapeutics and regulators on vascular morphology and branching tested in human clinical or laboratory animal experimental studies are quantified by comparing vascular parameters with control groups. VESGEN provides a user interface to both guide and allow control over the users vascular analysis process. An option is provided to select a morphological tissue type of vascular trees, network or tree-network composites, which determines the general collections of algorithms, intermediate images, and output images and measurements that will be produced.

Parsons-Wingerter, Patricia A.↗

A turbulent transport network model in MULTIFLUX coupled with TOUGH2

A new numerical method is described for the fully iterated, conjugate solution of two discrete submodels, involving (a) a transport network model for heat, moisture, and airflows in a high-permeability, air-filled cavity; and (b) a variably saturated fractured porous medium. The transport network submodel is an integrated-parameter, computational fluid dynamics solver, describing the thermal-hydrologic transport processes in the flow channel system of the cavity with laminar or turbulent flow and convective heat and mass transport, using MULTIFLUX. The porous medium submodel, using TOUGH2, is a solver for the heat and mass transport in the fractured rock mass. The new model solution extends the application fields of TOUGH2 by integrating it with turbulent flow and transport in a discrete flow network system. We present demonstrational results for a nuclear waste repository application at Yucca Mountain with the most realistic model assumptions and input parameters including the geometrical layout of the nuclear spent fuel and waste with variable heat load for the individual containers. The MULTIFLUX and TOUGH2 model elements are fully iterated, applying a programmed reprocessing of the Numerical Transport Code Functionalization model-element in an automated Outside Balance Iteration loop. The natural, convective airflow field and the heat and mass transport in a representative emplacement drift during postclosure are explicitly solved in the new model. The results demonstrate that the direction and magnitude of the air circulation patterns and all transport modes are strongly affected by the heat and moisture transport processes in the surrounding rock, justifying the need for a coupled, fully iterated model solution such as the one presented in the paper.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Addressing the Challenges of Anomaly Detection for Cyber Physical Energy Grid Systems

The consolidation of cyber communications networks and physical control systems within the energy smart grid introduces a number of new risks. Unfortunately, these risks are largely unknown and poorly understood, yet include very high impact losses from attack and component failures. One important aspect of risk management is the detection of anomalies and changes. However, anomaly detection within cyber security remains a difficult, open problem, with special challenges in dealing with false alert rates and heterogeneous data. Furthermore, the integration of cyber and physical dynamics is often intractable. And, because of their broad scope, energy grid cyber-physical systems must be analyzed at multiple scales, from individual components, up to network level dynamics. We describe an improved approach to anomaly detection that combines three important aspects. First, system dynamics are modeled using a reduced order model for greater computational tractability. Second, a probabilistic and principled approach to anomaly detection is adopted that allows for regulation of false alerts and comparison of anomalies across heterogeneous data sources. Third, a hierarchy of aggregations are constructed to support interactive and automated analyses of anomalies at multiple scales.

Anomaly Detection↗

Mitigate: An Adaptive Network Data Anonymization Tool Using Condensation-Based Differential Privacy

Modern network devices collect a large amount of data that can be analyzed to identify bottlenecks, anomalies, cyber-attacks, etc. Therefore, there is often a need to analyze such collections of network data quite often by an external expert or by the research community. However, these collections of data contain sensitive, proprietary information. In order for the network data to be shared, it must first be anonymized. The overall objective of this project is to develop an innovative privacy management tool to anonymize network data and achieve sufficient privacy, acceptable data utility, and efficient data analysis at the same time. No existing anonymization methods can achieve all of these at the same time. The core of this technology is a differential private clustering algorithm that provides strong privacy protection, preserves data properties important for subsequent analysis, and allows the party receiving the anonymized data to conduct analysis directly on anonymized data without the need of decryption or any extra processing. The research carried out was to design, implement and verify a solution to this problem by completing the following tasks: 1) developing the core technology; 2) developing a context based method that automatically recommends fields that must be anonymized; 3) conducted experiments showing superior results using our approach compared to existing tools, and 4) developed an intuitive but basic user interface. The research that was conducted generated novel algorithmic techniques that utilize state-of-the-art methods such as condensation, differential privacy preservation, clustering, automated tuning based on contextual awareness, and recommendation techniques to specify columns to users for anonymization leading to optimal privacy that allows research analysis on the dataset. Experiments were conducted to evaluate the efficacy of these novel algorithmic techniques by performing analysis on original non-anonymized datasets, then conducting analysis on the same yet anonymized datasets and comparing the results of the analyses. Overall, the anonymized analysis results were within 1% of the original results, verifying that the generated technology not only guarantees a high level of privacy but also enables research analysis as if it were conducted on the original dataset. Potential applications of this technology include anonymization of any type of structured network datasets that contain sensitive identifiers, such as IP addresses, that can be used in multiple applications. For example, to create an AI or machine learning model for cyber security, e.g., to detect attacks, or for performance analysis, e.g., identify bottlenecks or predict performance. In addition, a market analysis that was conducted for potential applications of this technology identified a broader range of applications of our anonymization technology beyond the network sector that includes healthcare, banking, insurance, securities, finance (FISB), data brokering, cloud services, ad sales, and government.

97 MATHEMATICS AND COMPUTING↗

NASA Tech Briefs, January 2012

Contents of this issue are: (1) Energy-Based Tetrahedron Sensor for High-Temperature, High-Pressure Environments (2) Handheld Universal Diagnostic Sensor (3) Large-Area Vacuum Ultraviolet Sensors (4) Fiber Bragg Grating Sensor System for Monitoring Smart Composite Aerospace Structures (5) Health-Enabled Smart Sensor Fusion Technology (6) Extended-Range Passive RFID and Sensor Tags (7) Hybrid Collaborative Learning for Classification and Clustering in Sensor Networks (8) Self-Healing, Inflatable, Rigidizable Shelter (9) Improvements in Cold-Plate Fabrication (10) Technique for Radiometer and Antenna Array Calibration - TRAAC (11) Real-Time Cognitive Computing Architecture for Data Fusion in a Dynamic Environment (12) Programmable Digital Controller (13) Use of CCSDS Packets Over SpaceWire to Control Hardware (14) Key Decision Record Creation and Approval Module (15) Enhanced Graphics for Extended Scale Range (16) Debris Examination Using Ballistic and Radar Integrated Software (17) Data Distribution System (DDS) and Solar Dynamic Observatory Ground Station (SDOGS) (18) Integration Manager (19) Eclipse-Free-Time Assessment Tool for IRIS (20) Automated and Manual Rocket Crater Measurement Software (21) MATLAB Stability and Control Toolbox Trim and Static Stability Module (22) Patched Conic Trajectory Code (23) Ring Image Analyzer (24) SureTrak Probability of Impact Display (25) Implementation of a Non-Metallic Barrier in an Electric Motor (26) Multi-Mission Radioisotope Thermoelectric Generator Heat Exchangers for the Mars Science Laboratory Rover (27) Uniform Dust Distributor for Testing Radiative Emittance of Dust-Coated Surfaces (28) MicroProbe Small Unmanned Aerial System (29) Highly Stable and Active Catalyst for Sabatier Reactions (30) Better Proton-Conducting Polymers for Fuel-Cell Membranes (31) CCD Camera Lens Interface for Real-Time Theodolite Alignment (32) Peregrine 100-km Sounding Rocket Project (33) SOFIA Closed- and Open-Door Aerodynamic Analyses (34) Sonic Thermometer for High-Altitude Balloons (35) Near-Infrared Photon-Counting Camera for High-Sensitivity Observations (36) Integrated Optics Achromatic Nuller for Stellar Interferometry (37) High-Speed Digital Interferometry (38) Ultra-Miniature Lidar Scanner for Launch Range Data Collection (39) Shape and Color Features for Object Recognition Search (40) Explanation Capabilities for Behavior-Based Robot Control (41) A DNA-Inspired Encryption Methodology for Secure, Mobile Ad Hoc Networks (42) Quality Control Method for a Micro-Nano-Channel Microfabricated Device (43) Corner-Cube Retroreflector Instrument for Advanced Lunar Laser Ranging (44) Electrospray Collection of Lunar Dust (45) Fabrication of a Kilopixel Array of Superconducting Microcalorimeters with Microstripline Wiring Spacecraft Attitude Tracking and Maneuver Using Combined Magnetic Actuators (46) Coherent Detector for Near-Angle Scattering and Polarization Characterization of Telescope Mirror Coatings

Source record↗

Performance evaluation of automated manufacturing systems using generalized stochastic Petri Nets

The main objective of this research is to develop a generic modeling methodology with a flexible and modular framework to aid in the design and performance evaluation of integrated manufacturing systems using a unified model. After a thorough examination of the available modeling methods, the Petri Net approach was adopted. The concurrent and asynchronous nature of manufacturing systems are easily captured by Petri Net models. Three basic modules were developed: machine, buffer, and Decision Making Unit. The machine and buffer modules are used for modeling transfer lines and production networks. The Decision Making Unit models the functions of a computer node in a complex Decision Making Unit Architecture. The underlying model is a Generalized Stochastic Petri Net (GSPN) that can be used for performance evaluation and structural analysis. GSPN's were chosen because they help manage the complexity of modeling large manufacturing systems. There is no need to enumerate all the possible states of the Markov Chain since they are automatically generated from the GSPN model.

Al-Jaar, Robert Y.↗

A hardware-in-the-loop (HIL) testbed for cyber-physical energy systems in smart commercial buildings

In recent years, there has been a growing trend toward the development of smart buildings that rely on cyber-physical systems (CPS) to optimize occupant comfort, safety, and energy efficiency. To ensure the reliable and efficient operation of CPS with designed control strategies, it is important to evaluate their performance under various scenarios before deploying them in the real world. This is where a Hardware-in-the-loop (HIL) testbed designed for studying sensor and control-related studies in smart buildings can be highly valuable. With the growing threat of cyber-attacks and physical faults targeting smart buildings, it is essential to ensure the security of building operations. A HIL testbed can emulate cyber-attack and physical fault scenarios, allowing researchers to develop and test threat detection and mitigation algorithms. This enables researchers to identify potential issues and optimize the algorithms in a safe and controlled environment before they are deployed in real-world settings, reducing the risk of failures that can negatively impact occupant comfort, safety, and energy efficiency. Therefore, this paper developed a HIL testbed designed for cyber-physical energy systems (e.g. buildings automation system (BAS)) in smart commercial buildings. The HIL testbed is comprised of a real-time building and Heating, Ventilation, and Air-Conditioning (HVAC) emulator using Modelica-based dynamic models, a set of BAS controllers, and a BAS computer server. The data generation capability of the HIL testbed is demonstrated by tracking normal and faulty operating data in the BAS, as well as monitoring detailed network traffic in the local BAS network. Here, this study further demonstrates the HIL testbed’s capability by conducting case studies on real-time physical fault and cyber-attack experiments using a Department of Energy (DOE) prototype commercial building. It is anticipated that the fully functional HIL testbed will be utilized for a variety of sensor and control-related studies, including but not limited to testing, developing, validating of different HVAC control strategies, fault detection & diagnosis, energy monitoring and analysis, cyber security study, etc.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Bridging Python to Silicon: The SODA Toolchain

Systems performing scientific computing, data analysis, and machine learning tasks have a growing demand for application-specific accelerators that can provide high computational performance while meeting strict size and power requirements. However, the algorithms and applications that need to be accelerated are evolving at a rate that is incompatible with manual design processes based on hardware description languages. Agile hardware design tools based on compiler techniques can help by quickly producing an application-specific integrated circuit (ASIC) accelerator starting from a high-level algorithmic description. Here, we present the software-defined accelerator (SODA) synthesizer, a modular and open-source hardware compiler that provides automated end-to-end synthesis from high-level software frameworks to ASIC implementation, relying on multilevel representations to progressively lower and optimize the input code. Our approach does not require the application developer to write any register-transfer level code, and it is able to reach up to 364 giga floating point operations per second (GFLOPS)/W efficiency (32-bit precision) on typical convolutional neural network operators.

97 MATHEMATICS AND COMPUTING↗