Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Computer Operations and Hardware”

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 703 records · Page 39

Optimization and Diagnostics Code for Technological Processes: Radiochemical Production Simulator

A software-hardware complex with the possibility of simulating the functioning of technological processes with APCS in real and accelerated time is being developed as part of Project Breakthrough in order to determine the optimal operating regimes of the technological processes of the units in a closed nuclear fuel cycle. It makes possible monitoring and control functions and the visualization and archiving of the results of simulation, which makes it possible at the development stages of technologies to evaluate the degree of adherence to the requirements of the technological equipment, regulating organs and final-control mechanisms, measurement apparatus, and monitoring channels, to analyze the control and diagnostics algorithms, blocking and protection, and to make when necessary changes in the design being developed. The developed software-hardware complex is a stimulator that will make it possible not only to investigate process schemes with their corresponding algorithms but also to develop training complexes for the training and certification of service personnel.

42 ENGINEERING↗

Dynamically adapting mechanism for translation lookaside buffer shootdowns

An operating system (OS) of a processing system having a plurality of processor cores determines a cost associated with different mechanisms for performing a translation lookaside buffer (TLB) shootdown in response to, for example, a virtual address being remapped to a new physical address, and selects a TLB shootdown mechanism to purge outdated or invalid address translations from the TLB based on the determined cost. In some embodiments, the OS selects an inter-processor interrupt (IPI) as the TLB shootdown mechanism if the cost associated with sending an IPI is less than a threshold cost. In some embodiments, the OS compares the cost of using an IPI as the TLB shootdown mechanism versus the cost of sending a hardware broadcast to all processor cores of the processing system as the shootdown mechanism and selects the shootdown mechanism having the lower cost.

42 ENGINEERING↗

RoboCount{sup TM} 2020 - A Robotic Non-Destructive Waste Assay System - 19331

A robotic system for non-destructive assay of radioactive waste - RoboCount{sup TM} 2020ab - has been designed, manufactured, and a prototype delivered to a customer. RoboCount{sup TM} 2020 is based on an industrial 6-axis articulating robot platform, combined with a dedicated waste package platform and a high-resolution HPGe gamma spectrometer. The system enables measurement of waste packages including drums (100-l, 200-l, 400-l), crates (up to 1200 x 800 x 800 mm), 1 m{sup 3} flexible intermediate bulk containers (FIBC), and others. Assay type can be freely defined, ranging from currently accepted protocols such as segmented gamma scan (SGS) or integral gamma scan (IGS) to new protocols developed for specific non-cylindrical waste packages, e.g., measurement of FIBCs from all four sides and the top. As a result of the robotic hardware flexibility, reconfiguration of the system for a new task is a matter of software changes only. In addition, while primarily geared towards waste characterization, the system can be configured for free release (clearance) by installing a second detector and field-of-view shielding. The prototype is currently in operation at a NPP site in Europe where it is used for D and D waste characterization in crates and FIBCs. Additional features beyond the basic functionality have been implemented such as an automatic lead collimator change based on measured HPGe detector dead time and an automatic liquid nitrogen refill, performed by the robot via a docking station at predefined time intervals. (Use of electrically cooled HPGe detectors is also possible). (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Quantum Information for Fusion Energy Sciences (Final Technical Report)

The simulation of plasma dynamics is a critical area of Fusion Energy Sciences (FES) due to it’s usefulness in predicting, controlling, and confining plasmas in the context of potential fusion reactors. The simulation of plasmas is a computationally difficult problem in both classical and quantum physics, motivating investigation into the potential of quantum computers to simulate these systems. This project took several concrete steps towards this goal by developing tools for improving the control, characterization, and calibration of quantum gates on a superconducting quantum computer, developing error suppression and mitigation tools to reduce errors on the quantum computer, and utilizing these advancements to simulate reduced models of plasma dynamics on the quantum computer. In order to efficiently simulate plasma physics, an optimal control method which synthesizes, directly at the pulse level, any quantum gate on qubit and qutrit systems was developed. Using four superconducting transmon quantum processors at Rigetti and LLNL, it was demonstrated that any arbitrary quantum gate on qubits and qutrits could be implemented with high fidelity, leading to a significantly reduced length of a gate sequence. A problem of interest in FES is the nonlinear optical process of laser pulse compression within a plasma. Since quantum physics is linear, simulating nonlinear operations is not naturally feasible on a quantum computer, however it is possible to simulated a quantized version of the nonlinear process. A quantization approach to convert nonlinear wave-wave interaction problems to Hamiltonian simulation problems was developed and demonstrated using two qubits on a Rigetti device. In this experiment, a number of error suppression and mitigation techniques were investigated to determine how best to utilize the finite quantum resources. This study provides an example of how plasma problems may be solved on near-term, noisy quantum computing platforms and identified a promising set of techniques. Building on the insights of these experiments, the investigation turned to linear electron-plasma wave physics. A connection was identified between a local one-dimensional lattice spin model and linear wave phenomena, allowing a plasma physics problem to be efficiently mapped to the quantum computer. In this framework, reflection and transmission of plasma waves at a sharp boundary was studied, as well as the propagation of waves through an inhomogeneous plasma medium. In addition to the suite of error suppression and mitigation techniques developed, this experiment introduced the use of a digital-analog gate scheme designed to efficiently simulate the plasma Hamiltonian. With hardware available at the conclusion of the project, simulation at the scale of 9 qubits and 15 timesteps (60 entangling layers) was achieved.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Keeping checkpoint/restart viable for exascale systems.

Next-generation exascale systems, those capable of performing a quintillion (10{sup 18}) operations per second, are expected to be delivered in the next 8-10 years. These systems, which will be 1,000 times faster than current systems, will be of unprecedented scale. As these systems continue to grow in size, faults will become increasingly common, even over the course of small calculations. Therefore, issues such as fault tolerance and reliability will limit application scalability. Current techniques to ensure progress across faults like checkpoint/restart, the dominant fault tolerance mechanism for the last 25 years, are increasingly problematic at the scales of future systems due to their excessive overheads. In this work, we evaluate a number of techniques to decrease the overhead of checkpoint/restart and keep this method viable for future exascale systems. More specifically, this work evaluates state-machine replication to dramatically increase the checkpoint interval (the time between successive checkpoint) and hash-based, probabilistic incremental checkpointing using graphics processing units to decrease the checkpoint commit time (the time to save one checkpoint). Using a combination of empirical analysis, modeling, and simulation, we study the costs and benefits of these approaches on a wide range of parameters. These results, which cover of number of high-performance computing capability workloads, different failure distributions, hardware mean time to failures, and I/O bandwidths, show the potential benefits of these techniques for meeting the reliability demands of future exascale platforms.

99 GENERAL AND MISCELLANEOUS↗

LEED: A Lightwave Energy-Efficient Datacenter

The Lightwave Energy-Efficient Datacenter (LEED) program is a disruptive “green-field” approach that provides a quantum leap in the energy efficiency of datacenters. LEED’s fundamental value proposition is that a novel and re-architected optical network—RotorNet— can deliver “more bandwidth per buck” as well as unique system-level attributes that significantly improve overall datacenter energy efficiency and performance. LEED has developed three system-level testbeds. The first testbed uses calibrated hardware and software power measurements to determine server energy efficiency as a function of network bandwidth and workload. These measurements have shown that increasing network communications bandwidth dramatically increases server energy efficiency providing a realistic path to the overall ENLITENED program goal of doubling the number of transactions per joule. The second testbed demonstrates key hardware: a prototype low-loss, high-port count optical “selector switch”. This switch was fabricated, racked, and tested. Measured switch characteristics include loss, bandwidth, crosstalk, switch time, system-level switch time (including the transceivers), and bit error rate. The third testbed demonstrates a fully working and manufactured pinwheel design which dramatically lowers the cost of design, while delivering high switch radix and low reconfiguration times. The LEED project has tied these three novel photonic switch prototypes together with production servers and software through the development of a novel FPGA-based NIC platform called Corundum. Corundum ensures that the packet-switched protocols supported by commodity operating systems and devices can interface with the Rotor switch design. The LEED group has used this combined hardware and software prototype to characterize applications running at a commercially relevant scale. The project has used a combination of enhanced optical modulation amplitude (OMA) modulators, broadband multiplexers and demultiplexers, avalanche photodiodes, and a novel burst-mode receivers to enable the insertion of LEED-developed optical switches without the need for expensive optical amplification. Our modeling has shown that measured LEED-developed device characteristics can achieve link characteristics of 2 pJ/bit including both transceivers and the Rotor switch. In summary, the LEED program has demonstrated a credible and practical path, through novel hardware and software, to realize the program objectives of ENLITENED. The net result will ensure that the United States maintains its strength in the crucial sector of Information Technology, which is vital to both our economic security and our national security.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

In-situ Condition Monitoring of Components in Small Modular Reactors Using Process and Electrical Signature Analysis. Final report, volume 1. Development of experimental flow control loop, data analysis and plant monitoring

The research and development under this project was focused on the following three major objectives: Objective 1: Identification of critical in-vessel SMR components for remote monitoring and development of their low-order dynamic models, along with a simulation model of an integral pressurized water reactor (iPWR). Objective 2: Development of an experimental flow control loop with motor-driven valves and pumps, incorporating data acquisition and on-line monitoring interface. Objective 3: Development of stationary and transient signal processing methods for electrical signatures, machinery vibration, and for characterizing process variables for equipment monitoring. This objective includes the development of a data analysis toolbox. The following is a summary of the technical accomplishments under this project: - A detailed literature review of various SMR types and electrical signature analysis of motor-driven systems was completed. A bibliography of literature is provided at the end of this report. Assistance was provided by ORNL in identifying some key references. - A review of literature on pump-motor modeling and digital signal processing methods was performed. - An existing flow control loop was upgraded with new instrumentation, data acquisition hardware and software. The upgrading of the experimental loop included the installation of a new submersible pump driven by a three-phase induction motor. All the sensors were calibrated before full-scale experimental runs were performed. - MATLAB-Simulink model of a three-phase induction motor and pump system was completed. The model was used to simulate normal operation and fault conditions in the motor-pump system, and to identify changes in the electrical signatures. - A simulation model of an integral PWR (iPWR) was updated and the MATLAB-Simulink model was validated for known transients. The pump-motor model was interfaced with the iPWR model for testing the impact of primary flow perturbations (upsets) on plant parameters and the pump electrical signatures. Additionally, the reactor simulation is being used to generate normal operation data and data with instrumentation faults and process anomalies. A frequency controller was interfaced with the motor power supply in order to vary the electrical supply frequency. The experimental flow control loop was used to generate operational data under varying motor performance characteristics. Coolant leakage events were simulated by varying the bypass loop flow rate. The accuracy of motor power calculation was improved by incorporating the power factor, computed from motor current and voltage in each phase of the induction motor.- A variety of experimental runs were made for steady-state and transient pump operating conditions. Process, vibration, and electrical signatures were measured using a submersible pump with variable supply frequency. High correlation was seen between motor current and pump discharge pressure signal; similar high correlation was exhibited between pump motor power and flow rate. Wide-band analysis indicated high coherence (in the frequency domain) between motor current and vibration signals. - Wide-band operational data from a PWR were acquired from AMS Corporation and used to develop time-series models, and to estimate signal spectrum and sensor time constant. All the data were from different pressure transmitters in the system, including primary and secondary loops. These signals were pre-processed using the wavelet transform for filtering both low-frequency and high-frequency bands. This technique of signal pre-processing provides minimum distortion of the data, and results in a more optimal estimation of time constants of plant sensors using time-series modeling techniques.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

Adaptive variational simulation for open quantum systems

Emerging quantum hardware provides new possibilities for quantum simulation. While much of the research has focused on simulating closed quantum systems, the real-world quantum systems are mostly open. Therefore, it is essential to develop quantum algorithms that can effectively simulate open quantum systems. Here we present an adaptive variational quantum algorithm for simulating open quantum system dynamics described by the Lindblad equation. The algorithm is designed to build resource-efficient ansatze through the dynamical addition of operators by maintaining the simulation accuracy. We validate the effectiveness of our algorithm on both noiseless simulators and IBM quantum processors and observe good quantitative and qualitative agreement with the exact solution. We also investigate the scaling of the required resources with system size and accuracy and find polynomial behavior. Our results demonstrate that near-future quantum processors are capable of simulating open quantum systems.

97 MATHEMATICS AND COMPUTING↗

Computed Axial Lithography (CAL):Toward Single Step 3D Printing of Arbitrary Geometries

Most additive manufacturing processes today operate by printing voxels (3D pixels) serially point-by-point to build up a 3D part. In some more recently-developed techniques, for example optical printing methods such as projection stereolithography [Zheng et al. 2012], [Tumbleston et al. 2015], parts are printed layer-by-layer by curing full 2d (very thin in one dimension) layers of the 3d part in each print step. There does not yet exist a technique which is able to print arbitrarily-defined 3D geometries in a single print step. If such a technique existed, it could be used to expand the range of printable geometries in additive manufacturing and relax constraints on factors such as overhangs in topology optimization. It could also vastly increase print speed for 3D parts. In this work, we develop the principles for an approach for single exposure 3D printing of arbitrarily defined geometries. Our approach, termed Computed Axial Lithgography (CAL), is based on tomographic reconstruction, with mathematical optimization to generate a set of projections to optically define an arbitrary dose distribution within a target volume. We demonstrate the potential ability of the technique to print 3D parts using a prototype CAL system based on sequential illumination from many angles. Finally, we propose new hardware designs which will help us to realize true single-shot arbitrary-geometry 3D CAL.

42 ENGINEERING↗

SU-F-T-462: Lessons Learned From a Machine Incident Reporting System

Purpose: Linear accelerators must operate with minimal downtime. Machine incident logs are a crucial tool to meet this requirement. They providing a history of service and demonstrate whether a fix is working. This study investigates the information content of a large department linear accelerator incident log. Methods: Our department uses an electronic reporting system to provide immediate information to both key department staff and the field service department. This study examines reports for five linac logs during 2015. The report attributes for analysis include frequency, level of documentation, who solved the problem, and type of fix used. Results: Of the reports, 36% were documented as resolved. In another 25% the resolution allowed treatment to proceed although the reported problem recurred within days. In 5% only intermediate troubleshooting was documented. The remainder lacked documentation. In 60% of the reports, radiation therapists resolved the problem, often by clearing the appropriate faults or reinitializing a software or hardware service. 22% were resolved by physics and 10% by field service engineers. The remaining 8% were resolved by IT, Facilities, or resolved spontaneously. Typical fixes, in order of scope, included clearing the fault and moving on, closing and re-opening the patient session or software, cycling power to a sub-unit, recalibrating a device (e.g., optical surface imaging), and calling in Field Service (usually resolving the problem through maintenance or component replacement). Conclusion: The reports with undocumented resolution represent a missed opportunity for learning. Frequency of who resolves a problem scales with the proximity of the person’s role (therapist, physicist, or service engineer), which is inversely related to the permanence of the resolution. Review of lessons learned from machine incident logs can form the basis for guidance to radiation therapists and medical physicists to minimize equipment downtime and ensure safe operation.

61 RADIATION PROTECTION AND DOSIMETRY↗

Current status of tritium calorimetry at TLK

Inside a tritium facility, calorimetry is an important analytical method as it is the only reference method for accountancy (it is based on the measurement of the heat generated by the radioactive decay). Presently, at Tritium Laboratory Karlsruhe (TLK), 4 calorimeters are in operation, one of isothermal type and three of inertial guidance control type (IGC). The volume of the calorimeters varies between 0.5 and 20.6 liters. About two years ago we started an extensive work to improve our calorimeters with regard to reliability and precision. We were forced to upgrade 3 of our 4 calorimeters due to the outdated interfaces and software. This work involved creating new LabView programs driving the devices, re-tuning control loops and replacing obsolete hardware components. In this paper we give a review on the current performance of our calorimeters, comparing it to recently available devices from the market and in the literature. We also show some ideas for a next generation calorimeter based on experiences with our IGC calorimeters and other devices reported in the literature. (authors)

07 ISOTOPE AND RADIATION SOURCES↗

The problem analysis of data processing performance in nuclear power plant computerized information and control system - 288

This paper is based on the data processing performance problems of operating command delay and system display not updating real-time for nuclear power plant computerized information and control system. The plant computerized information and control system is responsible for monitoring and controlling the plant operation, and it is a key part in the plant DCS system. Its data processing performance will affect the operation stability of nuclear power plant. The architecture, hardware configuration and data communication for plant computerized information and control system among the different levels in DCS are described. The main reasons affecting data processing performance in different signal channels are analyzed according to the plant load test results. The CPU load curves for real time server and central processing server are displayed, although the detailed cause for operating command delay and system display not updating realtime are not found, central processing server load is heavier than real time server which remind us to analyze the data processing of software resource scheduling deeply. So the resource distribution problem of software execution task is also analyzed according to CPU load test results. This paper provides the optimization analysis for data processing performance, and it is available and effective according to the latest test results. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Fail-Safe Logic Design Strategies Within Modern FPGA Architectures

Fail-safe computing refers to computing systems that revert to a non-operational safe state when a fault occurs. In this paper, we investigate a circuit level technique as mitigation for single event upsets (SEUs) and fault injection attacks on field programmable gate arrays (FPGAs), and analyze the effectiveness of the technique as a fail-safe monitor for an encryption algorithm. The propagation of fault effects through FPGA primitives including lookup tables (LUTs) and programmable interconnect points (PIPs) is assessed within an FPGA architecture created using an open source tool, and validated using fault injection experiments on an FPGA. The analysis reveals additional vulnerabilities exist within reconfigurable architectures over those in equivalent fail-safe application specific integrated circuit (ASIC), thus requiring a more elaborate network of redundant circuits and checking logic. The configuration memory bits (CMBs), which configure routing and designate logic functions within the LUTs of the FPGA, add complexity to fail-safe design strategies by introducing additional fault conditions and fault propagation paths. A resource-efficient fail-safe circuit design technique called DEsign for Fail-safe in reCONfigurable systems (DEFCON) is proposed. The benefits and limitations associated with DEFCON are described in the context of fault injection experiments carried out as simulations and in FPGA hardware.

Bhakta, Priya A. [Univ. of New Mexico, Albuquerque↗

Optimizing End-to-End Big Data Transfers over Terabits Network Infrastructure

While future terabit networks hold the promise of significantly improving big-data motion among geographically distributed data centers, significant challenges must be overcome even on today's 100 gigabit networks to realize end-to-end performance. Multiple bottlenecks exist along the end-to-end path from source to sink, for instance, the data storage infrastructure at both the source and sink and its interplay with the wide-area network are increasingly the bottleneck to achieving high performance. In this study, we identify the issues that lead to congestion on the path of an end-to-end data transfer in the terabit network environment, and we present a new bulk data movement framework for terabit networks, called LADS. LADS exploits the underlying storage layout at each endpoint to maximize throughput without negatively impacting the performance of shared storage resources for other users. LADS also uses the Common Communication Interface (CCI) in lieu of the sockets interface to benefit from hardware-level zero-copy, and operating system bypass capabilities when available. It can further improve data transfer performance under congestion on the end systems using buffering at the source using flash storage. With our evaluations, we show that LADS can avoid congested storage elements within the shared storage resource, improving input/output bandwidth, and data transfer rates across the high speed networks. We also investigate the performance degradation problems of LADS due to I/O contention on the parallel file system (PFS), when multiple LADS tools share the PFS. We design and evaluate a meta-scheduler to coordinate multiple I/O streams while sharing the PFS, to minimize the I/O contention on the PFS. Finally, with our evaluations, we observe that LADS with meta-scheduling can further improve the performance by up to 14 percent relative to LADS without meta-scheduling.

97 MATHEMATICS AND COMPUTING↗

Real-Time Health Monitoring for Gas Turbine Components Using Online Learning and High-Dimensional Data

Capital-intensive turbomachinery, such as gas turbines and combined cycle plants, are constantly being monitored for performance anomalies, faults, and physical degradation. Although these power-generating assets are equipped with hundreds of sensors, existing monitoring tools can only handle moderate-sized data. As a result, only a handful of aggregate metrics are used to monitor machine health. At the same time, developing advanced tools suitable for large datasets have been restricted by the lack of appropriate data. The objective of this proposal was to demonstrate a Big Data analytics framework for fault detection and diagnosis in gas turbine applications. We develop a predictive analytics framework methodology guided by these experimental data, industrial data from our collaborators, and physics-based models with engineering domain knowledge. Our analytics framework consists of four key components (1) a data curation process that addresses data storage, data quality assessments, and integrity checks, (2) a feature engineering component that utilizes statistical methods and transformation algorithms guided by physics-based models to extract high-fidelity fault features that can be leveraged for fault detection and classifying fault severities, (3) a Machine Learning-based fault detection and diagnostics algorithms for detecting operational and hardware faults in the combustion and the turbines section. We utilize two industry-class gas turbine component test rigs to generate first of its kind data for critical gas turbine faults with varying severity levels. Advanced gas turbine test facilities will be interrogated using state-of-the-art instrumentation techniques to build fault signatures and data trends for key combustor and turbine faults. Data generated from a combustor test rig (Georgia Tech) and a turbine test rig (Penn State) during both normal operation and with seeded faults serve as the basis for the Big Data sets. The test conditions in the two test facilities include common, critical events that occur in the operation. Utilizing the combustor test rig, we examine two common combustor faults: lean blowout and centerbody degradation. For the turbine section we develop analytic models for monitoring cooling faults in the gas turbine

03 NATURAL GAS↗

Data-Driven Digital Twin for Reliability Assessment of DC/DC Buck Converter

In commercial applications, the operation of DC/DC converters significantly impacts overall system performance and long-term reliability. This study introduces a data-driven digital twin (DT) approach for estimating critical degradation parameters of DC/DC BUCK converter under steady-state condition. Initially, a circuit-level MATLAB/Simulink digital model (DM C ) is refined against a hardware prototype’s switching model dataset using offline particle swarm optimization. The optimized digital model’s steady-state response is then verified with its average model response while varying the duty and load. Subsequently, degradation profiles are imposed on the inductor, capacitor, MOSFET in the DMC. A large dataset is generated from this model, allowing training, validation, and testing of machine learning (ML) models for component health regression tasks. The proposed method employs random forest ML models, achieving impressive regression results with a squared R value as high as 0.99978 and a root mean square error of 4.2× 10 –6 . The method is further validated on a medium power level DC/DC BUCK prototype with varying load conditions, and includes the analysis of MOSFET’s on-resistance under degradation conditions. This data-driven DT method shows promise for identifying parasitic degradation and ohmic loss parameters, enhancing converter reliability assessments in a non-invasive, generalized, and computationally efficient manner.

14 SOLAR ENERGY↗

An Evaluation of OpenSHMEM Interfaces for the Variable-Length Alltoallv() Collective Operation, In: OpenSHMEM and Related Technologies. Experiences, Implementations, and Technologies

Alltoallv() is a collective operation which allows all processes to exchange variable amounts of data with all other processes in the communication group. This means that Alltoallv() requires not only O(N2) communications, but typically also additional exchanges of the data lengths that will be transmitted in the eventual Alltoallv() call. This pre-exchange is used to calculate the proper offsets for the receiving buffers on the target processes. However, we propose two new candidate interfaces for Alltoallv() that would mitigate the need for the user to set up this extra exchange of information at the possible cost of memory efficiency. We explain the new interface variants and show how a single call can be used in place of the traditional Alltoall()/ Alltoallv() pair. We then discuss the performance tradeoffs for overall communication and memory costs, as well as both software and hardware-based optimizations and their applicability to the various proposed interfaces.

Computer Science↗

Sentinel

Network intrusion detection systems (NIDS) are commonplace in network security but they frequently employ algorithms that are computational demanding requiring hardware and software with significant power requirements. Two examples of such resource-intensive algorithms used for network security are regular expression matching and broader signature pattern matching which are commonly used in deep packet inspection (DPI). Network security algorithms that have large power requirements may be a challenge for low-power internet-of-things (IoT) environments, which generally lack the power resources to implement complex security measures like computationally expensive DPI at the edge. Furthermore, IoT environments incorporating 5G standalone networks have network latency constraints beyond just power that make DPI at the edge even more difficult. Programmable logic is ideally suited for machine learning inference for DPI because of its deep instruction level parallelism and single-cycle memory access. Machine learning approaches for DPI have been explored before using the programmable logic of field programmable gate arrays (FPGA) as a potential solution for NIDS approaches that would be power-suitable for IoT. However, those previous programmable logic NIDS approaches utilize either a supervised or unsupervised learning model. Sentinel utilizes the ensemble of these two machine learning approaches known as a semi-supervised approach which has shown promise in NIDS implementations. Sentinel provides a programmable logic implementation of a semi-supervised approach for DPI which operates at much lower power and latency than a GPU implementation with negligible loss of accuracy due to quantization through a logistic regressor.

Anderson, MatthewW [Idaho National Laboratory (INL↗