Efficient Performance Modeling Research Articles

Transfer Learning Enabled Modeling Paradigm for PVT-aware Circuit Performance Estimation

Designing robust performance models for modern complex digital circuits in the face of rapidly accelerating process variations is a critical yet demanding task. This paper introduces an efficient statistical performance modeling approach for VLSI digital circuits that incurs minimal computational expense. The fundamental concept involves capitalizing on knowledge gained from circuit modeling in one technology node to streamline the modeling process in another. This is achieved by merging previously established statistical models of process technology with a limited set of simulation data from a subsequent process technology through transfer learning. Comprehensive experiments conducted across diverse technology nodes demonstrate that the proposed framework is robust, precise, efficient in data usage, and computationally superior to other cutting-edge performance modeling techniques.

Data-driven machine learning approaches for simultaneous prediction of peak particle velocity and frequency induced by rock blasting in mining

The vibrations generated by rock blasting are a serious and hazardous outcome of these activities, causing harmful effects on the surrounding environment as well as the nearby residents. Both the local ecology and human communities suffer from the consequences of these vibrations. Assessing the severity of blasting vibrations necessitates a thorough evaluation of Peak Particle Velocity (PPV) and frequency, which are essential parameters for measuring vibration velocity. Accurate prediction of vibration occurrence is critically important. Therefore, this study employs five machine learning models for predicting the PPV and frequency resulting from quarry blasting. This work compares five machine learning models (XGBoost, Catboost, Bagging, Gradient Boosting, and Random Forest Regression) to choose the most efficient performance model. The performance evaluation of each five machine learning models demonstrates each model achieved a performance of more than 0.90 during the testing phase, there was a strong correlation observed between the actual and the predicted ones. The analysis of performance metrics shows Catboost regression model demonstrate better performance prediction comparing with the other models with R2 = 0.983, MSE = 0.000078, RMSE = 0.008, NRMSE = 0.019, MAD = 0.004, MAPE= 35.197 in the PPV prediction, and R2 = 0.975, MSE = 0.000243, RMSE = 0.015, NRMSE = 0.031, MAD = 0.008, MAPE = 37.281 for the frequency prediction. This study will help mining engineers and blasting experts to select the best machine learning model and its hyperparameters in estimating ground vibration, and frequency. In the context of the mining and civil industry, the application of this study offers significant potential for enhancing safety protocols and optimizing operational efficiency. By employing machine learning models, this research aims to accurately predict and assess ground vibrations with frequency resulting from rock blasting.

Predicting ground vibration during rock blasting using relevance vector machine improved with dual kernels and metaheuristic algorithms

The ground vibration caused by rock blasting is an extremely hazardous outcome of the blasting operation. Blasting activity has detrimental effects on both the ecology and the human population living in proximity to the area. Evaluating the magnitude of blasting vibrations requires careful evaluation of the peak particle velocity (PPV) as a fundamental and essential parameter for quantifying vibration velocity. Therefore, this study employs models using the relevance vector machine (RVM) approach for predicting the PPV resulting from quarry blasting. This investigation utilized the conventional and optimized RVM models for the first time in ground vibration prediction. This work compares thirty-three RVM models to choose the most efficient performance model. The following conclusions have been mapped from the outcomes of the several analyses. The performance evaluation of each RVM model demonstrates each model achieved a performance of more than 0.85 during the testing phase, there was a strong correlation observed between the actual ground vibrations and the predicted ones. The analysis of performance metrics (RMSE = 21.2999 mm/s, 16.2272 mm/s, R = 0.9175, PI = 1.59, IOA = 0.8239, IOS = 0.2541), score analysis (= 93), REC curve (= 6.85E−03, close to the actual, i.e., 0), curve fitting (= 1.05 close to best fit, i.e., 1), AD test (= 11.607 close to the actual, i.e., 9.790), Wilcoxon test (= 95%), Uncertainty analysis (WCB = 0.0134), and computational cost (= 0.0180) demonstrate that PSO_DRVM model MD29 outperformed better than other RVM models in the testing phase. This study will help mining and civil engineers and blasting experts to select the best kernel function and its hyperparameters in estimating ground vibration during rock blasting project. In the context of the mining and civil industry, the application of this study offers significant potential for enhancing safety protocols and optimizing operational efficiency.

Correlated Bayesian Model Fusion: Efficient High-Dimensional Performance Modeling of Analog/RF Integrated Circuits Over Multiple Corners

Efficient high-dimensional performance modeling of analog/RF circuits over multiple corners is an important-yet-challenging task. In this article, we propose a novel performance modeling approach for analog/RF circuits, referred to as correlated Bayesian model fusion (C-BMF). The key idea is to encode the correlation information for both model template and coefficient magnitude among different corners by using a unified prior distribution. Next, the prior distribution is combined with a few simulation samples via Bayesian inference to efficiently determine the unknown model coefficients. Two circuit examples designed in a commercial 40-nm CMOS process demonstrate that C-BMF achieves about <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$2\times $ </tex-math></inline-formula> cost reduction over the traditional state-of-the-art modeling technique without surrendering any accuracy.

DUET: A Novel Energy Yield Model With 3-D Shading for Bifacial Photovoltaic Systems

Bifacial photovoltaic (PV) performance models strive to accurately quantify rear-incident irradiance. While ray tracing models are optically rigorous, they require significant computational resources; faster view factor (VF) models are widely adopted but require user-defined loss factors to approximate rear shading and irradiance nonuniformity, introducing uncertainty in energy yield predictions. This article describes DUET—a bifacial PV performance model that calculates optical and electrical performance based on a physically representative array geometry. DUET’s novel shading algorithm pairs a 3-D VF model with deterministic ray-object intersections to capture 2-D shade-inclusive irradiance profiles while minimizing computational cost. Series and parallel combination of current–voltage curves capture irradiance nonuniformity throughout the module and array. This article provides validation against open-access system measurements from a test site in Roskilde, Denmark, and comparison to other software tested there [1]. DUET’s modeled bifacial energy yield agrees with measured data within −1.56% for fixed-tilt and −0.65% for horizontal single-axis tracked (HSAT) systems. Mean absolute error (MAE) in hourly bifacial power is 14.2–15.0 mW/Wp for fixed-tilt and 17.3–18.3 mW/Wp for HSAT, depending on the module temperature model applied. Comparing modeled and measured rear irradiance of two rear-facing pyranometers, DUET’s MAE values of 2.8 W/m <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$^{2}$</tex-math></inline-formula> for fixed-tilt and 3.7 W/m <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$^{2}$</tex-math></inline-formula> for HSAT are among the lowest errors reported for other software tested at this site. DUET provides computationally efficient bifacial performance modeling with geographic, temporal, and structural specificity to determine loss factors for use in other performance models or to be used directly in system design optimization.

Predictive Analysis of Diabetes-Risk with Class Imbalance.

Diabetes type 2 (T2DM) is a common chronic disease, increasingly leading to many complications and affecting vital organs. Hyperglycemia is the main characteristic caused by insufficient insulin secretion and poses a serious risk to human health. The objective is to construct a type-2 diabetes prediction model with high classification accuracy. Advanced machine learning and predictive model techniques are utilized to achieve cutting-edge techniques for the early diagnosis of diabetes. This paper proposes an efficient performance model to predict and classify the minority class of type-2 diabetes. The impact of oversampling and undersampling approaches to reduce the effect of an unbalanced class has been compared to classification performance algorithms. Synthetic Minority Oversampling (SMOTE) and Tomek-links techniques are applied and examined. The outcomes were then compared to the original unbalanced dataset using an artificial neural network (ANN) predictive model. The model is compared with other state-of-the-art classifiers such as support vector machine (SVM), random forest (RF), and decision tree (DT). The tuned model had the best accuracy of 92.2%. The experimental findings clearly manifest the improvement in accuracy and evaluation metrics in terms of AUC and F1-measure using the SMOTE oversampling strategy rather than the baseline and undersampling schemes. The study recommends adopting dynamic hyperparameter optimization to further improve accuracy.

Deterioration Pattern of Flexible Pavement with the Help of Falling Weight Deflectometer

Maintenance and repair of the highway network system are major expenses in the state budget. For this reason, various concerned organizations are pointing out the need for developing an intelligent and efficient pavement performance model that can prioritize pavement maintenance and rehabilitation works. Such models can forecast the remaining pavement service life and pavement rehabilitation needs, and can help in the formulation of pavement maintenance and strengthening programmes which will reduce the road agency and road user costs. The flexible pavement performance or deterioration models involve the complex interaction between vehicles, environment, structure and surface of the pavement. Performance models relating to the pavement distress conditions like, cracking, ravelling, potholing, and roughness are analysed and developed by various researchers. Understanding the deterioration pattern of the flexible pavement is very important in order to take the decision for strengthening the pavement. The remaining life of the pavement depends on various factors such as Traffic, Environment and climatic conditions hence keeping in mind these factors. the thesis presents the pattern of the deterioration of remaining life of pavement. The thesis emphasis on determining the remaining life of pavement by conducting the FWD test. The FWD test is conducted on the same pavement for three time at regular interval to verify the remaining life of the flexible pavement.

Development of accelerated PCHE off-design performance model for optimizing power system operation strategies in S-CO2 Brayton cycle

This paper presents the work on developing accurate and computationally efficient PCHE off-design quasi-steady state performance models for both recuperator and pre-cooler in a supercritical CO2 (S-CO2) Brayton cycle respectively to optimize power system operation strategies under off-design conditions. The newly developed model for heat transfer uses an existing Log Mean Temperature Difference (LMTD) method and modifies it to reflect the rapid change of properties in the vicinity of the critical point. In addition, the prediction of frictional pressure drop was implemented to this model. In the performance evaluation, this methodology shows its prediction accuracy to be sufficient. When it comes to the computational resource, the suggested model can accelerate the calculation over two orders of magnitude faster. The developed model is applied to evaluate the performance of the S-CO2 system to establish operation strategies such as inventory control and heater bypass control under off-design conditions.

Efficient Hierarchical Performance Modeling for Analog and Mixed-Signal Circuits via Bayesian Co-Learning

With the continuous drive toward integrated circuits scaling, efficient performance modeling is becoming more crucial yet more challenging. In this paper, we propose a novel method of hierarchical performance modeling based on Bayesian co-learning. We exploit the hierarchical structure of a circuit to establish a Bayesian framework where unlabeled data samples are generated to improve modeling accuracy without running additional simulation. Consequently, our proposed method only requires a small number of labeled samples, along with a large number of unlabeled samples obtained at almost no-cost, to accurately learn a performance model. Our numerical experiments demonstrate that the proposed approach achieves up to ${3.6 \times }$ runtime speed-up over the state-of-the-art modeling technique without surrendering any accuracy.

Efficient modeling of network flooding performance with proactive retransmissions in mobile ad hoc networks

Network flooding is the key mechanism designed to reach mobile nodes and disseminate the information in MANETs. Increasing the number of floods is a good way to improve coverage, namely, the fraction of nodes that receive the flooding message. However, this also will increase the overhead and hence the interference. In this paper, we used two different methods of refloods/retransmissions; originator-based retransmissions, where the originator of a message will retransmit the same message, and relay-based retransmissions, where each relay that has received the message will retransmit the same message. To compare two retransmission methods, we modeled the relationships of the coverage, overhead, and interference over the various flooding methods such as classic flooding method and several relay-set reduced flooding methods. Our analytic methods and numerical simulations made it possible to evaluate a wide range of scenarios (3,500,000 scenarios) with much less computational effort. In addition to analytic methods and numerical simulations, packet level simulations were performed to justify the results obtained from our efficient performance models. Furthermore, we investigated the impact of proactive retransmissions on the overhead and coverage as a function of message generation rate. When sequentially considering one transmission, originator-based optimal retransmission, and relay-based optimal retransmission, CF coverage becomes lower and lower than efficient flooding methods at the higher node density and faster message generation rate.

Bayesian Model Fusion: Large-Scale Performance Modeling of Analog and Mixed-Signal Circuits by Reusing Early-Stage Data

Efficient performance modeling of today’s analog and mixed-signal circuits is an important yet challenging task, due to the high-dimensional variation space and expensive circuit simulation. In this paper, we propose a novel performance modeling algorithm that is referred to as Bayesian model fusion (BMF) to address this challenge. The key idea of BMF is to borrow the information collected from an early stage (e.g., schematic level) to facilitate efficient performance modeling at a late stage (e.g., post layout). Such a goal is achieved by statistically modeling the performance correlation between early and late stages through Bayesian inference. Furthermore, to make the proposed BMF method of practical utility, four implementation issues, including: 1) prior mapping; 2) missing prior knowledge; 3) fast solver; and 4) prior and hyper-parameter selection, are carefully considered in this paper. Two circuit examples designed in a commercial 32 nm CMOS silicon on insulator process demonstrate that the proposed BMF method achieves up to $9\times $ runtime speed-up over the traditional modeling technique without surrendering any accuracy.

Institutional conformance and tourism performance: an efficiency analysis in developing Pacific Island countries

ABSTRACTUsing a multi-level approach that exploits the synergism between hospitality and tourism, this study examines conceptual and empirical linkages between institutional isomorphism and tourism performance in developing Pacific Island countries. To this end, data envelopment analysis is used to examine how coercive, normative, and mimetic conformity mechanisms are related to optimally efficient tourism performance models. The results provide useful insights for policymakers and development strategists at the macro level by identifying the tourism models that do, and do not, lead to such efficiency. At the micro level, the results can help tourism and hospitality operators move toward more judicious uses of resources in environments characterized by both constrained tourism development potential and constrained rationality. Finally, research on institutional isomorphism in tourism and hospitality is scarce, and thus this study advances an interesting theoretical lens for future research in this area.

Efficient Hybrid Performance Modeling for Analog Circuits Using Hierarchical Shrinkage Priors

Efficient performance modeling is an extremely important task for yield analysis and design optimization of analog circuits. In this paper, a novel regression modeling method based on hierarchical shrinkage priors is proposed to construct hybrid performance models with both high accuracy and low computational cost. In particular, the user-defined model templates derived from design equations and the general-purpose orthogonal polynomials are combined together to set up a hybrid dictionary. Next, in order to avoid over-shrinking large model coefficients, a novel regression method based on hierarchical shrinkage priors and variational Bayesian inference is adopted for model fitting. A rail-to-rail operational amplifier example demonstrates that the proposed method achieves up to 40% error reduction over other state-of-the-art approaches without increasing the modeling cost.

A new approach for signal and noise FET modeling including wave propagation effects

AbstractIn the millimeter‐wave frequency range, electromagnetic effects can significantly influence a device behavior. As the core of modern communication systems, active devices such as field‐effect transistors require up‐to‐date models to accurately integrate such effects, especially in terms of noise performance because most of communication systems operate in noisy environments. Based on the active coupled line concept, a novel approach for efficient noise performance modeling of millimeter‐wave field‐effect transistors is proposed. This distributed model considers the effect of wave propagation along the device electrodes, which can significantly affect the noise performance, especially in the millimeter‐wave range. Copyright © 2015 John Wiley &amp; Sons, Ltd.

Efficient and accurate analytical performance models for translucent optical networks

This paper presents new analytical models for computing the blocking performance of translucent optical networks with sparsely located optical-electrical-optical regeneration nodes. Due to physical layer impairments such as noise and crosstalk, there is a maximum distance [called the transmission reach (TR)] that optical signals can travel before they have to be regenerated. Regeneration nodes that are allocated to a lightpath can also be used for wavelength conversion besides mitigating PLIs. The quality of transmission of a lightpath is satisfied if the length of each segment between two allocated regeneration nodes is less than the TR. We develop analytical models to estimate the call blocking probability for two typical regeneration node allocation policies. The first policy allocates regeneration nodes when the lightpath needs either regeneration or wavelength conversion, while the other allocation policy allocates regenerators only for signal regeneration. The analytical models strike a balance between computational complexity and prediction accuracy. The models are validated using extensive simulation results for a variety of network topologies and parameters.

Efficient and Accurate Analytical Performance Models for Translucent Optical Networks

Efficient and Accurate Analytical Performance Models for Translucent Optical Networks

Critical review of flexible pavement performance models

Maintenance and repair of the highway network system are major expenses in the state budget. For this reason various concerned organizations are pointing out the need for developing an intelligent and efficient pavement performance model that can prioritize pavement maintenance and rehabilitation works. Such models can forecast the remaining pavement service life and pavement rehabilitation needs, and can help in the formulation of pavement maintenance and strengthening programmes which will reduce the road agency and road user costs. The flexible pavement performance or deterioration models involve the complex interaction between vehicles and the environment, and the structure and surface of the pavement. Performance models relating to the pavement distress conditions like, cracking, raveling, potholing, and roughness are analyzed and developed by various researchers. But most of these models are found applicable to a particular set of traffic or environment conditions, thus highlighting the need of model(s) that can work in varied conditions satisfactorily. The paper presents a detailed review of various pavement performance models to examine the role of factors related to pavement materials, environmental conditions, type of traffic and volume of traffic, and to identify the limitations and gaps in the present knowledge on such models.

Exploiting task and data parallelism for advanced video coding on hybrid CPU + GPU platforms

Considering the prevalent usage of multimedia applications on commodity computers equipped with both CPU and GPU devices, the possibility of simultaneously exploiting all parallelization capabilities of such hybrid platforms for high performance video encoding has been highly quested for. Accordingly, a method to concurrently implement the H.264/ advanced video coding (AVC) inter-loop on hybrid GPU + CPU platforms is proposed in this manuscript. This method comprises dynamic dependency aware task distribution methods and real-time computational load balancing over both the CPU and the GPU, according to an efficient dynamic performance modeling. With such optimal balance, the set of rather optimized parallel algorithms that were conceived for video coding on both the CPU and the GPU are dynamically instantiated in any of the existing processing devices, to minimize the overall encoding time. The proposed model does not only provide an efficient task scheduling and load balancing for H.264/AVC inter-loop, but it also does not introduce any significant computational burden to the time-limited video coding application. Furthermore, according to the presented set of experimental results, the proposed scheme has proved to provide speedup values as high as 2.5 when compared with highly optimized GPU-only encoding solutions or even other state of the art algorithm. Moreover, by simply using the existing computational resources that usually equip most commodity computers the proposed scheme is able to achieve inter-loop encoding rates as high as 40 fps at a HD 1920 × 1080 resolution.

Introduction to the theme issue on performance modeling

The increasing complexity of modern computer systems demands new and efficient performance modeling techniques to aid developers in producing the most efficient system for a desired task. Performance evaluation seeks to understand and explain system performance. It should be an integral component of the development process from the beginning to reduce the costs required to fix problems late in the development cycle. Indeed, decisions made early in the process will directly impact the quality of the final product. The model-based development paradigm, in contrast, advocates developing systems starting from models—at multiple levels of abstraction—that express domain-specific concepts precisely and intuitively while supporting automated manipulation and transformation. The goal of this theme issue is to illuminate the deep relationship between computer systems performance evaluation and system modeling. The papers presented in this issue help to answer such questions as—How can model-driven techniques that are used to understand a system’s behavior be used to improve its performance? Can modeling languages and tools be adapted for improving performance? What are the connections between modeling tool environments and performance evaluation tools and environments? The papers in this theme issue are a step towards bridging the gap between the theory and practice of performance evaluation and system modeling.

Loss Performance Modeling for Hierarchical Heterogeneous Wireless Networks With Speed-Sensitive Call Admission Control

A hierarchical overlay structure is an alternative solution that integrates existing and future heterogeneous wireless networks to provide subscribers with better mobile broadband services. Traffic loss performance in such integrated heterogeneous networks is necessary for an operator's network dimensioning and planning. This paper investigates the computationally efficient loss performance modeling for multiservice in hierarchical heterogeneous wireless networks. A speed-sensitive call admission control (CAC) scheme is considered in our model to assign overflowed calls to appropriate tiers. This approach avoids unnecessary and frequent handoff between cells and reduces signaling overheads. An approximation model with guaranteed accuracy and low computational complexity is presented for the loss performance of multiservice traffic. The accuracy of numerical results is validated by comparing the results from the approximation with simulations.