SEARCH · Search NASA
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
We present low-cost and effective methods and results for high levels of Total Ionizing Dose testing performed on the Xilinx Kintex UltraScale (20nm) Field-Programmable Gate Array evaluation board using 60Co and 2 MeV electron sources.
Offloading computation from a CPU to a hardware accelerator is becoming a more common solution for improving performance because traditional gains enabled by Moore’s law and Dennard scaling have slowed. GPUs are often used as hardware accelerators, but field-programmable gate arrays (FPGAs) are gaining traction. FPGAs are beneficial because they allow hardware specific to a particular application to be created. However, they are notoriously difficult to program. To this end, two of the main FPGA manufacturers, Intel and Xilinx, have created tools and frameworks that enable the use of higher level languages to design FPGA hardware. Although Xilinx kernels can be designed by using C/C++, both Intel and Xilinx support the use of OpenCL C to architect FPGA hardware. However, not much is known about the portability and performance between these two device families other than the fact that it is theoretically possible to synthesize a kernel meant for Intel to Xilinx and vice versa.In this work, we evaluate the portability and performance of Intel and Xilinx kernels. We use OpenCL C implementations of a subset of the Rodinia benchmarking suite that were designed for an Intel FPGA and make the necessary modifications to create synthesizable OpenCL C kernels for a Xilinx FPGA. We find that the difficulty of porting certain kernel optimizations varies, depending on the construct. Once the minimum amount of modifications is made to create synthesizable hardware for the Xilinx platform, more nontrivial work is needed to improve performance. However, we find that constructs that are known to be performant for an FPGA should improve performance regardless of the platform; the difficulty comes in deciding how to invoke certain kernel optimizations while also abiding by the constraints enforced by a given platform’s hardware compiler.
In this article, we present an end-to-end workflow for superconducting qubit readout that embeds codesigned neural networks into the quantum instrumentation control kit (QICK). Capitalizing on the custom firmware and software of the QICK platform, which is built on Xilinx radiofrequency system-on-chip field-programmable gate arrays (FPGAs), we aim to leverage machine learning (ML) to address critical challenges in qubit readout accuracy and scalability. The workflow utilizes the hls4ml package and employs quantization-aware training to translate ML models into hardware-efficient FPGA implementations via user-friendly Python application programming interfaces. We experimentally demonstrate the design, optimization, and integration of an ML algorithm for single transmon qubit readout, achieving 96% single-shot fidelity with a latency of 32.25 ns and less than 16% FPGA lookup table resource utilization. Our results offer the community an accessible workflow to advance ML-driven readout and adaptive control in quantum information processing applications.
Wireless sensors connected in a local network offer revolutionary exploration capabilities, but the current solutions do not work in extreme environments of low temperatures (200K) and low to moderate radiation levels (<50 krad). These sensors (temperature, radiation, infrared, etc.) would need to operate outside the spacecraft/ lander and be totally independent of power from the spacecraft/lander. Flash memory field-programmable gate arrays (FPGAs) are being used as the main signal processing and protocol generation platform in a new receiver. Flash-based FPGAs have been shown to have at least 100 reduced standby power and 10 reduction operating power when compared to normal SRAM-based FPGA technology.
Recent advancements in use of machine learning (ML) techniques on field-programmable gate arrays (FPGAs) have allowed for the implementation of embedded neural networks with extremely low latency. This is invaluable for particle detectors at the Large Hadron Collider, where latency and used area are strictly bounded. The hls4ml framework is a procedure that converts trained ML model software to a synthesis result to can be used on an FPGA. However, running the pipeline is a time-consuming procedure, and there is a strong risk of failure. In particular, it may not be possible to successfully convert a model into a synthesis result, or the resource consumption of the model may exceed the resources of the target FPGA. To aid with this development, we introduce wa-hls4ml, a surrogate model using a graph neural network to emulate the structure of the source models. The goal is to estimate the chance of success and resource consumption of a given model when passed through the hls4ml pipeline, without needing to run the pipeline.
We introduce an automated tool for deploying ultra low-latency, low-power deep neural networks with convolutional layers on field-programmable gate arrays (FPGAs). By extending the hls4ml library, we demonstrate an inference latency of 5 µs using convolutional architectures, targeting microsecond latency applications like those at the CERN Large Hadron Collider. Considering benchmark models trained on the Street View House Numbers Dataset, we demonstrate various methods for model compression in order to fit the computational constraints of a typical FPGA device used in trigger and data acquisition systems of particle detectors. In particular, we discuss pruning and quantization-aware training, and demonstrate how resource utilization can be significantly reduced with little to no loss in model accuracy. We show that the FPGA critical resource consumption can be reduced by 97% with zero loss in model accuracy, and by 99% when tolerating a 6% accuracy degradation.
Recent advancements in use of machine learning techniques on field-programmable gate arrays (FPGAs) have allowed for implementation of embedded neural networks with extremely low latency. This is invaluable for particle detectors at the Large Hadron Collider, where latency and used area must be strictly bounded. The hls4ml framework is a procedure for converting from trained machine learning model software, to a synthesis result that can be used on an FPGA. However, running the pipeline is a time-consuming procedure, and there is a strong risk of failure. In particular, it is possible that the model is unable to be converted into a synthesis result, or that the resource consumption of the model will exceed the resources of the target FPGA. To aid with this development, we introduce wa-hls4ml, a surrogate model which uses a graph neural network to emulate the structure of the source models. The goal is to estimate the chance of success and resource consumption of an arbitrary model when passed through the hls4ml procedure, without the time consumption of actually running the pipeline.
Techniques and methodologies are developed for assuring and verifying FPGAs that adequately address the increased complexity of the devices required by today'\'s missions.
Abstract Recurrent neural networks have been shown to be effective architectures for many tasks in high energy physics, and thus have been widely adopted. Their use in low-latency environments has, however, been limited as a result of the difficulties of implementing recurrent architectures on field-programmable gate arrays (FPGAs). In this paper we present an implementation of two types of recurrent neural network layers—long short-term memory and gated recurrent unit—within the hls4ml framework. We demonstrate that our implementation is capable of producing effective designs for both small and large models, and can be customized to meet specific design requirements for inference latencies and FPGA resources. We show the performance and synthesized designs for multiple neural networks, many of which are trained specifically for jet identification tasks at the CERN Large Hadron Collider.
High-performance digital electronic circuits for onboard processing of return signals in an airborne precipitation- measuring radar system have been implemented in commercially available field-programmable gate arrays (FPGAs). Previously, it was standard practice to downlink the radar-return data to a ground station for postprocessing a costly practice that prevents the nearly-real-time use of the data for automated targeting. In principle, the onboard processing could be performed by a system of about 20 personal- computer-type microprocessors; relative to such a system, the present FPGA-based processor is much smaller and consumes much less power. Alternatively, the onboard processing could be performed by an application-specific integrated circuit (ASIC), but in comparison with an ASIC implementation, the present FPGA implementation offers the advantages of (1) greater flexibility for research applications like the present one and (2) lower cost in the small production volumes typical of research applications. The generation and processing of signals in the airborne precipitation measuring radar system in question involves the following especially notable steps: The system utilizes a total of four channels two carrier frequencies and two polarizations at each frequency. The system uses pulse compression: that is, the transmitted pulse is spread out in time and the received echo of the pulse is processed with a matched filter to despread it. The return signal is band-limited and digitally demodulated to a complex baseband signal that, for each pulse, comprises a large number of samples. Each complex pair of samples (denoted a range gate in radar terminology) is associated with a numerical index that corresponds to a specific time offset from the beginning of the radar pulse, so that each such pair represents the energy reflected from a specific range. This energy and the average echo power are computed. The phase of each range bin is compared to the previous echo by complex conjugate multiplication to obtain the mean Doppler shift (and hence the mean and variance of the velocity of precipitation) of the echo at that range.
Progress has been made in a continuing effort to develop a spaceborne computer system for processing readout data from a Fourier-transform infrared (FTIR) spectrometer to reduce the volume of data transmitted to Earth. The approach followed in this effort, oriented toward reducing design time and reducing the size and weight of the spectrometer electronics, has been to exploit the versatility of recently developed hybrid field-programmable gate arrays (FPGAs) to run diverse software on embedded processors while also taking advantage of the reconfigurable hardware resources of the FPGAs.
A product line of high-definition color video cameras, now under development, offers a superior combination of desirable characteristics, including high frame rates, high resolutions, low power consumption, and compactness. Several of the cameras feature a 3,840 2,160-pixel format with progressive scanning at 30 frames per second. The power consumption of one of these cameras is about 25 W. The size of the camera, excluding the lens assembly, is 2 by 5 by 7 in. (about 5.1 by 12.7 by 17.8 cm). The aforementioned desirable characteristics are attained at relatively low cost, largely by utilizing digital processing in advanced field-programmable gate arrays (FPGAs) to perform all of the many functions (for example, color balance and contrast adjustments) of a professional color video camera. The processing is programmed in VHDL so that application-specific integrated circuits (ASICs) can be fabricated directly from the program. ["VHDL" signifies VHSIC Hardware Description Language C, a computing language used by the United States Department of Defense for describing, designing, and simulating very-high-speed integrated circuits (VHSICs).] The image-sensor and FPGA clock frequencies in these cameras have generally been much higher than those used in video cameras designed and manufactured elsewhere. Frequently, the outputs of these cameras are converted to other video-camera formats by use of pre- and post-filters.
Field-programmable gate arrays (FPGAs) are becoming more promising for high performance computing (HPC). Unlike fixed hardware specialization, reconfigurability allows FPGAs to be customized for a wide variety of HPC workloads. Our ultimate goal is to understand the viability of FPGAs as accelerators for some of the key HPC codes at LLNL and identify application characteristics that can benefit from the reconfigurability of FPGAs. More specifically, this LDRD project conducted a detailed performance study of FPGAs using sparse matrix computations and identified key optimization techniques to effectively use FPGA hardware resources.
The figure is a schematic diagram of a complementary metal oxide/semiconductor (CMOS) electronic circuit that performs one of three different logic functions, depending on the level of an externally applied control voltage, V(sub sel). Specifically, the circuit acts as A NAND gate at V(sub sel) = 0.0 V, A wire (the output equals one of the inputs) at V(sub sel) = 1.0 V, or An AND gate at V(sub sel) = -1.8 V. [The nominal power-supply potential (VDD) and logic "1" potential of this circuit is 1.8 V.] Like other multifunctional circuits described in several prior NASA Tech Briefs articles, this circuit was synthesized following an automated evolutionary approach that is so named because it is modeled partly after the repetitive trial-and-error process of biological evolution. An evolved circuit can be tested by computational simulation and/or tested in real hardware, and the results of the test can provide guidance for refining the design through further iteration. The evolutionary synthesis of electronic circuits can now be implemented by means of a software package Genetic Algorithms for Circuit Synthesis (GACS) that was developed specifically for this purpose. GACS was used to synthesize the present trifunctional circuit. As in the cases of other multifunctional circuits described in several prior NASA Tech Briefs articles, the multiple functionality of this circuit, the use of a single control voltage to select the function, and the automated evolutionary approach to synthesis all contribute synergistically to a combination of features that are potentially advantageous for the further development of robust, multiple-function logic circuits, including, especially, field-programmable gate arrays (FPGAs). These advantages include the following: This circuit contains only 9 transistors about half the number of transistors that would be needed to obtain equivalent NAND/wire/AND functionality by use of components from a standard digital design library. If multifunctional gates like this circuit were used in the place of the configurable logic blocks of present commercial FPGAs, it would be possible to change the functions of the resulting digital systems within shorter times. For example, by changing a single control voltage, one could change the function of thousands of FPGA cells within nanoseconds. In contrast, typically, the reconfiguration in a conventional FPGA by use of bits downloaded from look-up tables via a digital bus takes microseconds.
A document discusses the replication of the functionality of the onboard space-shuttle general-purpose computers (GPCs) in field-programmable gate arrays (FPGAs) and application-specific integrated circuits (ASICs). The purpose of the replication effort is to enable utilization of proven space-shuttle flight software and software-development facilities to the extent possible during development of software for flight computers for a new generation of launch vehicles derived from the space shuttles. The replication involves specifying the instruction set of the central processing unit and the input/output processor (IOP) of the space-shuttle GPC in a hardware description language (HDL). The HDL is synthesized to form a "core" processor in an FPGA or, less preferably, in an ASIC. The core processor can be used to create a flight-control card to be inserted into a new avionics computer. The IOP of the GPC as implemented in the core processor could be designed to support data-bus protocols other than that of a multiplexer interface adapter (MIA) used in the space shuttle. Hence, a computer containing the core processor could be tailored to communicate via the space-shuttle GPC bus and/or one or more other buses.
The Adaptive Instrument Module (AIM) will be the first true demonstration of reconfigurable computing with field-programmable gate arrays (FPGAs) in space, enabling the 'brain' of the system to evolve or adapt to changing requirements. In partnership with NASA Goddard Space Flight Center and the Australian Cooperative Research Centre for Satellite Systems (CRC-SS), APL has built the flight version to be flown on the Australian university-class satellite FEDSAT. The AIM provides satellites the flexibility to adapt to changing mission requirements by reconfiguring standardized processing hardware rather than incurring the large costs associated with new builds. This ability to reconfigure the processing in response to changing mission needs leads to true evolveable computing, wherein the instrument 'brain' can learn from new science data in order to perform state-of-the-art data processing. The development of the AIM is significant in its enormous potential to reduce total life-cycle costs for future space exploration missions. The advent of RAM-based FPGAs whose configuration can be changed at any time has enabled the development of the AIM for processing tasks that could not be performed in software. The use of the AIM enables reconfiguration of the FPGA circuitry while the spacecraft is in flight, with many accompanying advantages. The AIM demonstrates the practicalities of using reconfigurable computing hardware devices by conducting a series of designed experiments. These include the demonstration of implementing data compression, data filtering, and communication message processing and inter-experiment data computation. The second generation is the Adaptive Processing Template (ADAPT) which is further described in this paper. The next step forward is to make the hardware itself adaptable and the ADAPT pursues this challenge by developing a reconfigurable module that will be capable of functioning efficiently in various applications. ADAPT will take advantage of radiation tolerant RAM-based field programmable gate array (FPGA) technology to develop a reconfigurable processor that combines the flexibility of a general purpose processor running software with the performance of application specific processing hardware for a variety of high performance computing applications.
High-speed, low-power, reconfigurable electronic hardware has been developed to implement ICER-3D, an algorithm for compressing hyperspectral-image data. The algorithm and parts thereof have been the topics of several NASA Tech Briefs articles, including Context Modeler for Wavelet Compression of Hyperspectral Images (NPO-43239) and ICER-3D Hyperspectral Image Compression Software (NPO-43238), which appear elsewhere in this issue of NASA Tech Briefs. As described in more detail in those articles, the algorithm includes three main subalgorithms: one for computing wavelet transforms, one for context modeling, and one for entropy encoding. For the purpose of designing the hardware, these subalgorithms are treated as modules to be implemented efficiently in field-programmable gate arrays (FPGAs). The design takes advantage of industry- standard, commercially available FPGAs. The implementation targets the Xilinx Virtex II pro architecture, which has embedded PowerPC processor cores with flexible on-chip bus architecture. It incorporates an efficient parallel and pipelined architecture to compress the three-dimensional image data. The design provides for internal buffering to minimize intensive input/output operations while making efficient use of offchip memory. The design is scalable in that the subalgorithms are implemented as independent hardware modules that can be combined in parallel to increase throughput. The on-chip processor manages the overall operation of the compression system, including execution of the top-level control functions as well as scheduling, initiating, and monitoring processes. The design prototype has been demonstrated to be capable of compressing hyperspectral data at a rate of 4.5 megasamples per second at a conservative clock frequency of 50 MHz, with a potential for substantially greater throughput at a higher clock frequency. The power consumption of the prototype is less than 6.5 W. The reconfigurability (by means of reprogramming) of the FPGAs makes it possible to effectively alter the design to some extent to satisfy different requirements without adding hardware. The implementation could be easily propagated to future FPGA generations and/or to custom application-specific integrated circuits.
With the discovery of the Higgs boson at the CERN LHC, the world's highest-energy particle accelerator complex, scientists have acquired an important tool to study the fundamental building blocks of the universe. Precision measurements of Higgs bosons produced with large momentum allow for unique insights into the structure of the interactions of the Higgs boson with other particles that may shed light on physics beyond the standard model. While experimentally challenging, exploring such interactions with novel artificial intelligence (AI) methods can advance our understanding of the Higgs sector, including the Higgs boson's self-interaction. Moreover, the LHC is undergoing a major upgrade to further increase its particle collision rate and thereby operate for an additional decade. The experimental detectors at the upgraded facility must process at least a factor of ten more data at rates of hundreds of terabytes per second all under challenging conditions. New AI techniques are required to reconstruct and select, or trigger on, the most physics-sensitive events in real-time to handle the resulting avalanche of data. The proposed research will achieve the goals of the LHC program at the CMS experiment by developing a sub-microsecond event reconstruction system using real-time AI algorithms that employ field-programmable gate array technologies. By harnessing sophisticated AI techniques, this research focuses on measuring the production of Higgs bosons at large momentum while enhancing particle reconstruction methods in the trigger and beyond. Overall, the proposed research has broader implications for the use of AI in resource-constrained, low-latency embedded applications across all fields of science.