Search NASA⌕ Search

SEARCH · Search NASA

Results for “timing error evaluation”

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 145 records · Page 8

Development of real-time rotating waveplate Stokes polarimeter using multi-order retardation for ITER poloidal polarimeter

The rotating waveplate Stokes polarimeter was developed for ITER (International Thermonuclear Experimental Reactor) poloidal polarimeter. The generalized model of the rotating waveplate Stokes polarimeter and the algorithm suitable for real-time field-programmable gate array (FPGA) processing were proposed. Since the generalized model takes into account each component associated with the rotation of the waveplate, the Stokes parameters can be accurately measured even in unideal condition such as non-uniformity of the waveplate retardation. Experiments using a He-Ne laser showed that the maximum error and the precision of the Stokes parameter were 3.5% and 1.2%, respectively. The rotation speed of waveplate was 20 000 rpm and time resolution of measuring the Stokes parameter was 3.3 ms. Software emulation showed that the real-time measurement of the Stokes parameter with time resolution of less than 10 ms is possible by using several FPGA boards. Evaluation of measurement capability using a far-infrared laser which ITER poloidal polarimeter will use concluded that measurement error will be reduced by a factor of nine.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Molecular radiotherapy: The NUKFIT software for calculating the time-integrated activity coefficient

Purpose: Calculation of the time-integrated activity coefficient (residence time) is a crucial step in dosimetry for molecular radiotherapy. However, available software is deficient in that it is either not tailored for the use in molecular radiotherapy and/or does not include all required estimation methods. The aim of this work was therefore the development and programming of an algorithm which allows for an objective and reproducible determination of the time-integrated activity coefficient and its standard error.Methods: The algorithm includes the selection of a set of fitting functions from predefined sums of exponentials and the choice of an error model for the used data. To estimate the values of the adjustable parameters an objective function, depending on the data, the parameters of the error model, the fitting function and (if required and available) Bayesian information, is minimized. To increase reproducibility and user-friendliness the starting values are automatically determined using a combination of curve stripping and random search. Visual inspection, the coefficient of determination, the standard error of the fitted parameters, and the correlation matrix are provided to evaluate the quality of the fit. The functions which are most supported by the data are determined using the corrected Akaike information criterion. The time-integrated activity coefficient is estimated by analytically integrating the fitted functions. Its standard error is determined assuming Gaussian error propagation. The software was implemented using MATLAB.Results: To validate the proper implementation of the objective function and the fit functions, the results of NUKFIT and SAAM numerical, a commercially available software tool, were compared. The automatic search for starting values was successfully tested for reproducibility. The quality criteria applied in conjunction with the Akaike information criterion allowed the selection of suitable functions. Function fit parameters and their standard error estimated by using SAAM numerical and NUKFIT showed differences of <1%. The differences for the time-integrated activity coefficients were also <1% (standard error between 0.4% and 3%). In general, the application of the software is user-friendly and the results are mathematically correct and reproducible. An application of NUKFIT is presented for three different clinical examples.Conclusions: The software tool with its underlying methodology can be employed to objectively and reproducibly estimate the time integrated activity coefficient and its standard error for most time activity data in molecular radiotherapy.

61 RADIATION PROTECTION AND DOSIMETRY↗

Development and Implementation of Dynamic Scripts to Support Local Model Verification at National Weather Service Weather Forecast Offices

Local modeling with a customized configuration is conducted at National Weather Service (NWS) Weather Forecast Offices (WFOs) to produce high-resolution numerical forecasts that can better simulate local weather phenomena and complement larger scale global and regional models. The advent of the Environmental Modeling System (EMS), which provides a pre-compiled version of the Weather Research and Forecasting (WRF) model and wrapper Perl scripts, has enabled forecasters to easily configure and execute the WRF model on local workstations. NWS WFOs often use EMS output to help in forecasting highly localized, mesoscale features such as convective initiation, the timing and inland extent of lake effect snow bands, lake and sea breezes, and topographically-modified winds. However, quantitatively evaluating model performance to determine errors and biases still proves to be one of the challenges in running a local model. Developed at the National Center for Atmospheric Research (NCAR), the Model Evaluation Tools (MET) verification software makes performing these types of quantitative analyses easier, but operational forecasters do not generally have time to familiarize themselves with navigating the sometimes complex configurations associated with the MET tools. To assist forecasters in running a subset of MET programs and capabilities, the Short-term Prediction Research and Transition (SPoRT) Center has developed and transitioned a set of dynamic, easily configurable Perl scripts to collaborating NWS WFOs. The objective of these scripts is to provide SPoRT collaborating partners in the NWS with the ability to evaluate the skill of their local EMS model runs in near real time with little prior knowledge of the MET package. The ultimate goal is to make these verification scripts available to the broader NWS community in a future version of the EMS software. This paper provides an overview of the SPoRT MET scripts, instructions for how the scripts are run, and example use cases.

Zavordsky, Bradley↗

Development and Implementation of Dynamic Scripts to Support Local Model Verification at National Weather Service Weather Forecast Offices

Local modeling with a customized configuration is conducted at National Weather Service (NWS) Weather Forecast Offices (WFOs) to produce high-resolution numerical forecasts that can better simulate local weather phenomena and complement larger scale global and regional models. The advent of the Environmental Modeling System (EMS), which provides a pre-compiled version of the Weather Research and Forecasting (WRF) model and wrapper Perl scripts, has enabled forecasters to easily configure and execute the WRF model on local workstations. NWS WFOs often use EMS output to help in forecasting highly localized, mesoscale features such as convective initiation, the timing and inland extent of lake effect snow bands, lake and sea breezes, and topographically-modified winds. However, quantitatively evaluating model performance to determine errors and biases still proves to be one of the challenges in running a local model. Developed at the National Center for Atmospheric Research (NCAR), the Model Evaluation Tools (MET) verification software makes performing these types of quantitative analyses easier, but operational forecasters do not generally have time to familiarize themselves with navigating the sometimes complex configurations associated with the MET tools. To assist forecasters in running a subset of MET programs and capabilities, the Short-term Prediction Research and Transition (SPoRT) Center has developed and transitioned a set of dynamic, easily configurable Perl scripts to collaborating NWS WFOs. The objective of these scripts is to provide SPoRT collaborating partners in the NWS with the ability to evaluate the skill of their local EMS model runs in near real time with little prior knowledge of the MET package. The ultimate goal is to make these verification scripts available to the broader NWS community in a future version of the EMS software. This paper provides an overview of the SPoRT MET scripts, instructions for how the scripts are run, and example use cases.

Zavodsky, Bradley↗

A passive submicrosecond time dissemination system

A submicrosecond time dissemination system was developed. The system is composed of an experimental transit improvement satellite (TRIAD), TRANET Satellite Tracking System, and special satellite signal receivers each tied to separate cesium clocks. The 400 MHz signal from the satellite is phase modulated plus or minus 45 deg by a pseudo-random noise (PRN) digital code with a bit rate of 5/3 MHz. Evaluation of this system has shown that satellite oscillator frequency instabilities are the major source of error. Experiments indicate that Global time transfers can be made with errors less than 100 nanoseconds.

Taylor, R. J.↗

Evaluation Analysis of NASA SMAP L3 and L4 and SPoRT-LIS Soil Moisture Data in the United States

Soil moisture has a critical role in the development, frequency and persistence of climatic and hydrologic extremes such as drought, heat wave and flooding events. In situ soil moisture data are uneven and sparse in time and space. This highlights the need to utilize other soil moisture sources to fill this spatiotemporal gap. The goal of this study is to validate one satellite-based and two model-based soil moisture datasets with in situ data across the United States. Soil moisture information from the Soil Moisture Active Passive (SMAP) enhanced level 3 (L3) (SMAP L3) and modeled level 4 (L4) (SMAP L4) data at 9-km resolution and Short-term Prediction Research and Transition-Land Information System (SPoRT-LIS) modeled at 3-km resolution were selected for evaluation. SPoRT-LIS is a near real-time, high resolution operational land analysis data. Ground-based data were obtained from the North American Soil Moisture Database (NASMD) for 362 stations. Seven statistical indicators including anomaly, Spearman, and Pearson correlation coefficients, the systematic error (Bias), root mean square error (RMSE), unbiased root mean square error (ubRMSE), and normalized standard deviation (SDV) were used to evaluate the satellite- and model-derived soil moisture data. In addition, the triple collocation (TC) error model was used to measure the error among SMAP L4, SPoRT-LIS and ground-based data. This study assesses which satellite or modeled dataset is most appropriate for specific times and locations to use as a surrogate for in situ observations. Temporal and spatial analysis demonstrated that, overall, SMAP L4 performed better than SMAP L3 and SPoRT-LIS. Strong agreement was observed between SMAP L4 and in situ observations (ρ = 0.53, Bias = −0.006) in all seasons and most regions with various land covers, especially in winter and in the central regions of the United States. For croplands, SMAP L4 presented the best agreement with in situ data, analyzing all period (ρ = 0.60) and non-winter period (ρ = 0.61) separately.

Tavakol, Ameneh↗

Using an external surrogate for predictor model training in real-time motion management of lung tumors

Purpose: Precise prediction of respiratory motion is a prerequisite for real-time motion compensation techniques such as beam, dynamic couch, or dynamic multileaf collimator tracking. Collection of tumor motion data to train the prediction model is required for most algorithms. To avoid exposure of patients to additional dose from imaging during this procedure, the feasibility of training a linear respiratory motion prediction model with an external surrogate signal is investigated and its performance benchmarked against training the model with tumor positions directly. Methods: The authors implement a lung tumor motion prediction algorithm based on linear ridge regression that is suitable to overcome system latencies up to about 300 ms. Its performance is investigated on a data set of 91 patient breathing trajectories recorded from fiducial marker tracking during radiotherapy delivery to the lung of ten patients. The expected 3D geometric error is quantified as a function of predictor lookahead time, signal sampling frequency and history vector length. Additionally, adaptive model retraining is evaluated, i.e., repeatedly updating the prediction model after initial training. Training length for this is gradually increased with incoming (internal) data availability. To assess practical feasibility model calculation times as well as various minimum data lengths for retraining are evaluated. Relative performance of model training with external surrogate motion data versus tumor motion data is evaluated. However, an internal–external motion correlation model is not utilized, i.e., prediction is solely driven by internal motion in both cases. Results: Similar prediction performance was achieved for training the model with external surrogate data versus internal (tumor motion) data. Adaptive model retraining can substantially boost performance in the case of external surrogate training while it has little impact for training with internal motion data. A minimum adaptive retraining data length of 8 s and history vector length of 3 s achieve maximal performance. Sampling frequency appears to have little impact on performance confirming previously published work. By using the linear predictor, a relative geometric 3D error reduction of about 50% was achieved (using adaptive retraining, a history vector length of 3 s and with results averaged over all investigated lookahead times and signal sampling frequencies). The absolute mean error could be reduced from (2.0 ± 1.6) mm when using no prediction at all to (0.9 ± 0.8) mm and (1.0 ± 0.9) mm when using the predictor trained with internal tumor motion training data and external surrogate motion training data, respectively (for a typical lookahead time of 250 ms and sampling frequency of 15 Hz). Conclusions: A linear prediction model can reduce latency induced tracking errors by an average of about 50% in real-time image guided radiotherapy systems with system latencies of up to 300 ms. Training a linear model for lung tumor motion prediction with an external surrogate signal alone is feasible and results in similar performance as training with (internal) tumor motion. Particularly for scenarios where motion data are extracted from fluoroscopic imaging with ionizing radiation, this may alleviate the need for additional imaging dose during the collection of model training data.

60 APPLIED LIFE SCIENCES↗

Efficient and generalizable nested Fourier-DeepONet for three-dimensional geological carbon sequestration

Geological carbon sequestration (GCS) involves injecting CO2 into subsurface geological formationsfor permanent storage. Numerical simulations could guide decisions in GCS projects by predictingCO 2 migration pathways and the pressure distribution in storage formation. However, these simula-tions are often computationally expensive due to highly coupled physics and large spatial-temporalsimulation domains. Surrogate modelling with data-driven machine learning has become a promis-ing alternative to accelerate physics-based simulations. Among these, the Fourier neural operator(FNO) has been applied to three-dimensional synthetic subsurface models. Despite its good accuracyin simulating CO 2 plume migration, it requires large computational resources in training and alsolacks generalizability. Here, to further improve performance, we have developed a nested Fourier-DeepONet by combining the expressiveness of the FNO with the modularity of a deep operatornetwork (DeepONet). This new framework is twice as efficient as a nested FNO for training and has atleast 80% lower GPU memory requirement due to its flexibility to treat temporal coordinates sepa-rately. These performance improvements are achieved without compromising prediction accuracy.In addition, the generalization and extrapolation ability of nested Fourier-DeepONet beyond thetraining range has been thoroughly evaluated. Nested Fourier-DeepONet outperformed the nestedFNO for extrapolation in time with more than 50% reduced error. It also exhibited good extrapolationaccuracy beyond the training range in terms of reservoir properties, number of wells, and injectionrate.

Lee, Jonathan E. [Department of Chemical and Envir↗

Multilevel Conditional Disturbance Rejection Control for Satellite Attitude Tracking

Recently, the conditional disturbance rejection controller (CDRC) was proposed to improve control performance by leveraging disturbances that have beneficial effects on the system. However, it only considers disturbances acting on the state variable directly influenced by them. Although fast convergence of this state can be achieved with the CDRC, it may unintentionally affect the convergence of the output (i.e., the primary state). Here, in this article, a multilevel CDRC is proposed to enhance satellite attitude control performance by accounting for the effect of disturbances on both attitude (output) and angular velocity. The extended state observer is employed to estimate the lumped disturbance, including the modeling errors and external disturbances. Then, a backstepping-based controller with the multilevel disturbance rejection law (ML-DRL) is designed for attitude tracking. The ML-DRL is developed to improve the control performance by using a disturbance with a damping effect on both attitude and velocity. Faster convergence of attitude and velocity can be achieved by conditionally compensating for the disturbance. The stability of the proposed control method is analyzed by demonstrating that the errors are bounded as time tends to infinity. The attitude control performance of the proposed method is evaluated through numerical examples conducted using the MATLAB/Simulink Multibody tool.

Active disturbance rejection control (ADRC)↗

Dynamic Response of a Planetary Gear System Using a Finite Element/Contact Mechanics Model

The dynamic response of a helicopter planetary gear system is examined over a wide range of operating speeds and torques. The analysis tool is a unique, semianalytical finite element formulation that admits precise representation of the tooth geometry and contact forces that are crucial in gear dynamics. Importantly, no a priori specification of static transmission error excitation or mesh frequency variation is required; the dynamic contact forces are evaluated internally at each time step. The calculated response shows classical resonances when a harmonic of mesh frequency coincides with a natural frequency. However, peculiar behavior occurs where resonances expected to be excited at a given speed are absent. This absence of particular modes is explained by analytical relationships that depend on the planetary configuration and mesh frequency harmonic. The torque sensitivity of the dynamic response is examined and compared to static analyses. Rotation mode response is shown to be more sensitive to input torque than translational mode response.

Parker, Robert G.↗

Required Time of Arrival as a Control Mechanism to Mitigate Uncertainty in Arrival Traffic Demand Management

The objective of this study is to explore the use of Required Time of Arrival (RTA) capability on the flight deck as a control mechanism on arrival traffic management to improve traffic delivery accuracy by mitigating the effect of traffic demand uncertainty. The uncertainties are caused by various factors, such as departure error due to the difference between scheduled departure and the actual take-off time. A simulation study was conducted using the Multi Aircraft Control System (MACS) software, a comprehensive research platform developed in the Airspace Operations Laboratory (AOL) at NASA Ames Research Center. The Crossing Time (CT) performance (i.e. the difference between target crossing time and actual crossing time) of the RTA for uncertainty mitigation during cruise phase was evaluated under the influence of varying two main factors: wind severity (heavy wind vs. mild wind), and wind error (1 hour, 2 hours, and 5 hours wind forecast errors). To examine the CT performance improvement made by the RTA, the comparison to the CT of the aircraft that were not assigned with RTA (Non-RTA) under the influence of the selected factors was also made. The Newark Liberty International Airport (EWR) was chosen for this study. A total 66 inbound traffic to the EWR (34 of them were airborne when the simulation was initiated, 32 were pre-departures at that time) was simulated, where the pre-scripted departure error was assigned to each pre-departure (61 conform to their Expected Departure Clearance Time, which is +-300 seconds of their scheduled departure time). The results of the study show that the delivery accuracy improvement can be achieved by assigning RTA, regardless of the influence of the selected two factors (the wind severity and the wind information inaccuracy). Across all wind variances, 66.9 (265 out of 396) of the CT performance of the RTA assigned aircraft was within +- 60 seconds (i.e. target tolerance range) and 88.9 (352 out of 396) aircraft met +-300 seconds marginal tolerance range, while only 33.6 (133 out of 396) of the Non-RTA assigned aircrafts CT performance achieved the target tolerance range and 75.5 (299 out of 396) stayed within the marginal. Examination of the impact of different error sources i.e. departure error, wind severity, and wind error suggest that although large departure errors can significantly impact the CT performance, the impacts of wind severity and errors were modest relative the targeted +- 60 second conformance range.

required time of arrival (RTA)↗

Global Horizontal Irradiance Clear Sky Models: Implementation and Analysis

Clear sky models estimate the terrestrial solar radiation under a cloudless sky as a function of the solar elevation angle, site altitude, aerosol concentration, water vapor, and various atmospheric conditions. This report provides an overview of a number of global horizontal irradiance (GHI) clear sky models from very simple to complex. Validation of clear-sky models requires comparison of model results to measured irradiance during clear-sky periods. To facilitate validation, we present a new algorithm for automatically identifying clear-sky periods in a time series of GHI measurements. We evaluate the performance of selected clear-sky models using measured data from 30 different sites, totaling about 300 site-years of data. We analyze the variation of these errors across time and location. In terms of error averaged over all locations and times, we found that complex models that correctly account for all the atmospheric parameters are slightly more accurate than other models, but, primarily at low elevations, comparable accuracy can be obtained from some simpler models. However, simpler models often exhibit errors that vary with time of day and season, whereas the errors for complex models vary less over time.

99 GENERAL AND MISCELLANEOUS↗

An advanced teleoperator control system - Design and evaluation

The design goal of an advanced teleoperator control system is twofold: 1) to allow the operator's manual control to be robust to system nonlinearities such as time delays and operator's control errors, and 2) to support the high performance of teleoperation while reducing the operator's control burden by providing the master and slave arms with desirable dynamic properties and by allowing the slave arm to automatically perform such control tasks as compliance and force control in the form of task sharing. The authors present a novel teleoperator control system achieving the above design goal by taking the following into consideration: the human dynamics involved in generating control command based on visual and forced feedback is modeled and incorporated into the controller design and evaluation; the dynamic characteristics of slave and master arms are actively modified in such a way as to implement the desirable dynamic characteristics; and the force feedback is redefined in terms of the combination of opposition and force discrepancies in order to establish the required man/machine dynamic coordination under shared control. The proposed control system with human dynamics in the control loop is simulated and compared with a number of conventional methods in the presence of human control errors and time delays.

Lee, Sukhan↗

Simplified formula for mean cycle-slip time of phase-locked loops with steady-state phase error.

Previous work shows that the mean time from lock to a slipped cycle of a phase-locked loop is given by a certain double integral. Accurate numerical evaluation of this formula for the second-order loop is extremely vexing because the difference between exponentially large quantities is involved. The presented article demonstrates a method in which a much-reduced precision program can be used to obtain the mean first-cycle slip time for a loop of arbitrary degree tracking at a specified SNR and steady-state phase error. It also presents a simple approximate formula that is asymptotically tight at higher loop SNR.

Tausworthe, R. C.↗

Metamodeling of Droplet Activation for Global Climate Models

The nucleation of cloud droplets from the ambient aerosol is a critical physical process that must be resolved for global models to faithfully predict aerosol–cloud interactions and aerosol indirect effects on climate. To better represent droplet nucleation from a complex, multimodal, and multicomponent aerosol population within the context of a global model, a new metamodeling framework is applied to derive an efficient and accurate activation parameterization. The framework applies polynomial chaos expansion to a detailed parcel model in order to derive an emulator that maps thermodynamic and aerosol parameters to the supersaturation maximum achieved in an adiabatically ascending parcel and can be used to diagnose droplet number from a single lognormal aerosol mode. The emulator requires much less computational time to build, store, and evaluate than a high-dimensional lookup table. Compared to large sample sets from the detailed parcel model, the relative error in the predicted supersaturation maximum and activated droplet number computed with the best emulator is and (one standard deviation), respectively. On average, the emulators constructed here are as accurate and between 10 and 17 times faster than a leading physically based activation parameterization. Because the underlying parcel model being emulated resolves size-dependent droplet growth factors, the emulator captures kinetic limitations on activation. Furthermore, the results discussed in this work suggest that this metamodeling framework can be extended to accurately account for the detailed activation of a complex aerosol population in an arbitrary coupled global aerosol–climate model.

54 ENVIRONMENTAL SCIENCES↗

SU-E-T-473: A Patient-Specific QC Paradigm Based On Trajectory Log Files and DICOM Plan Files

Purpose: To evaluate a remote QC tool for monitoring treatment machine parameters and treatment workflow. Methods: The Varian TrueBeamTM linear accelerator is a digital machine that records machine axis parameters and MLC leaf positions as a function of delivered monitor unit or control point. This information is saved to a binary trajectory log file for every treatment or imaging field in the patient treatment session. A MATLAB analysis routine was developed to parse the trajectory log files for a given patient, compare the expected versus actual machine and MLC positions as well as perform a cross-comparison with the DICOM-RT plan file exported from the treatment planning system. The parsing routine sorts the trajectory log files based on the time and date stamp and generates a sequential report file listing treatment parameters and provides a match relative to the DICOM-RT plan file. Results: The trajectory log parsing-routine was compared against a standard record and verify listing for patients undergoing initial IMRT dosimetry verification and weekly and final chart QC. The complete treatment course was independently verified for 10 patients of varying treatment site and a total of 1267 treatment fields were evaluated including pre-treatment imaging fields where applicable. In the context of IMRT plan verification, eight prostate SBRT plans with 4-arcs per plan were evaluated based on expected versus actual machine axis parameters. The average value for the maximum RMS MLC error was 0.067±0.001mm and 0.066±0.002mm for leaf bank A and B respectively. Conclusion: A real-time QC analysis program was tested using trajectory log files and DICOM-RT plan files. The parsing routine is efficient and able to evaluate all relevant machine axis parameters during a patient treatment course including MLC leaf positions and table positions at time of image acquisition and during treatment.

60 APPLIED LIFE SCIENCES↗

A simplified procedure for correcting both errors and erasures of a Reed-Solomon code using the Euclidean algorithm

It is well known that the Euclidean algorithm or its equivalent, continued fractions, can be used to find the error locator polynomial and the error evaluator polynomial in Berlekamp's key equation needed to decode a Reed-Solomon (RS) code. A simplified procedure is developed and proved to correct erasures as well as errors by replacing the initial condition of the Euclidean algorithm by the erasure locator polynomial and the Forney syndrome polynomial. By this means, the errata locator polynomial and the errata evaluator polynomial can be obtained, simultaneously and simply, by the Euclidean algorithm only. With this improved technique the complexity of time domain RS decoders for correcting both errors and erasures is reduced substantially from previous approaches. As a consequence, decoders for correcting both errors and erasures of RS codes can be made more modular, regular, simple, and naturally suitable for both VLSI and software implementation. An example illustrating this modified decoding procedure is given for a (15, 9) RS code.

Truong, T. K.↗

Dynamic Undervolting to Improve Energy Efficiency on Multicore X86 CPUs

Chip manufacturers introduce redundancy at various levels of CPU design to guarantee correct operation, even for worst-case combinations of non-idealities in process variation and system operating conditions. This redundancy is implemented partly in the form of voltage margins. However, for a wide range of real-world execution scenarios these margins are excessive and merely translate to increased power consumption, hindering the effort towards higher-energy efficiency in both HPC and general purpose computing. Our study on the x86-64 Haswell and Skylake multicore microarchitectures reveals-wide voltage margins, which vary across different microarchitectures, different chip parts of the same microarchitecture, and across different workloads. We find that it is necessary to quantify-voltage margins using multi-threaded and multi-instance workloads, as characterization with single-threaded and single-instance workloads that do not stress the CPU to its full capacity typically identifies overly optimistic margins that lead to errors when applied in realistic program execution scenarios. In addition, we introduce, deploy and evaluate a run-time governor that dynamically reduces the supply voltage of modern multicore x86-64 CPUs. Our governor employs a model that takes as input a set of performance metrics which are directly measurable via performance monitoring counters and have high predictive value for the minimum tolerable supply voltage (V min ), to predict and apply the appropriate reduction for the workload at hand. Compared with the conventional DVFS governor, our approach in this study achieves up to 42 percent energy savings for the Skylake family and 34 percent for the Haswell family for complex, real-world applications.

97 MATHEMATICS AND COMPUTING↗