Deriving Power Models for Architecture-Level Energy Efficiency Analyses

MuMMI (Multiple Metrics Modeling Infrastructure) environment is an infrastructure that facilitates systematic measurement, modeling, and prediction of performance, power consumption and performance-power tradeoffs for parallel systems. MuMMI builds upon three existing frameworks: Prophesy for performance modeling and prediction of parallel applications, PAPI for hardware performance counter monitoring, and PowerPack for power measurement and profiling. In this paper, we present the MuMMI framework, which consists of an Instrumentor, Databases and Analyzer. The MuMMI Instrumentor provides automatic performance and power data collection and storage with low overhead. The MuMMI Databases extend the databases of Prophesy to store power and energy consumption and hardware performance counters' data with different CPU frequency settings. The MuMMI Analyzer extends the data analysis component of Prophesy to support power consumption and hardware performance counters, and it entails performance and power modeling, performance-power tradeoff and optimizations, and web-based automated modeling system. Currently, our MuMMI online automated performance and power modeling system uses four modeling techniques: curve fitting, parameterization, kernel coupling and performance-counters-based, we discuss the effort to automate the process of developing performance and power models for scientific applications online, and focus on exploring performance-counters-based performance and power modeling. The MuMMI environment is able to aid in performance and power data measurement, storage, modeling and prediction of scientific applications on XSEDE resources in XSEDE community.

Due to the increasing demand of cloud resources, the ever-increasing number and scale of cloud data centers make their massive power consumption a prominent issue today. Evidence reveals that the behaviors of cloud servers make the major impact on data centers’ power consumption. Although extensive research can be found in this context, a systematic review of the models and modeling methods for the entire hierarchy (from underlying hardware components to the upper-layer applications) of the cloud server is still missing, which is supposed to cover the relevant studies on physical and virtual cloud server instances, server components, and cloud applications. In this article, we summarize a broad range of relevant studies from three perspectives: power data acquisition, power models, and power modeling methods for cloud servers (including bare-metal, virtual machine (VM), and container instances). We present a comprehensive taxonomy on the collection methods of server-level power data, the existing mainstream power models at multiple levels from hardware to software and application, and commonly used methods for modeling power consumption including classical regression analysis and emerging methods like reinforcement learning. Throughout the work, we introduce a variety of models and methods, illustrating their implementation, usability, and applicability while discussing the limitations of existing approaches and possible ways of improvement. Apart from reviewing existing studies on server power models and modeling methods, we further figure out several open challenges and possible research directions, such as the study on modeling the power consumption of lightweight virtual units like unikernel and the necessity of further explorations toward empowering server power estimation/prediction with machine learning. As power monitoring is drawing increasing attention from cloud service providers (CSPs), this survey provides useful guidelines on server power modeling and can be inspiring for further research on energy-efficient data centers.

In order to support orders of traffic explosion with sustainable energy consumption, besides conventional Spectrum Efficiency (SE), Energy Efficiency (EE) is proposed as another critical metric for the next generation mobile networks (5G). This paper proposes to evaluate the potential EE gain of 5G on the basis of EARTH Energy Efficiency Evaluation Framework (E3F). As E3F indicated, the large-scale network traffic over the long term is definitely heterogeneous, which is still the trend in 5G. In this paper, relationship is deduced between the global EE metric and the snapshot EE metrics, which are obtained from small scale and short term. To be specific, this paper considers two typical 5G scenarios as different snapshots: the indoor office and the outdoor open space gathering. Each snapshot scenario has its own combinations of service/applications, hence distinct traffic model and resulted network deployment, power consumption and EE. The power model used for EE evaluation considers the different bandwidth and antenna configurations for different types of Base Stations (BSs) in different scenarios, meanwhile, it identifies the long term hardware evolution as a key factor to predict future power consumptions. Numerical analysis reveals that up to 1000x EE gain can be achieved by 5G and gain of hundreds times can be realized in 5G era, when 4G coexist with 5G.

As power consumption results in greenhouse gas emissions and energy costs for operators, analyzing power consumption in wireless networks and portable devices is of crutial importance. Due to environmental effects resulted from energy generation and exploitation as well as the cost of surging energy, energy-aware wireless systems attract unprecedented attention. Cognitive Radio (CR) is one of the optimal solutions that allows for energy savings on both the networks and devices. Thus, cognitive radio contributes to increase spectral and energy efficiency as well as reduction in power consumption. In addition, energy consumption of the CR technologies as intelligent technology should be considered to realize the green networks objective. In this article, we look into energy efficiency of the cognitive wireless network paradigms. Moreover, energy efficiency analysis and modelling in these systems are specifically focused on achieving green communications objectives. However, CRs by altering all elements of wireless data communications are considered in this paper, and the energy-efficient operation and energy efficiency enabler perspectives of CRs are also analyzed.

Study on Energy and Environmental Efficiency for Coal-fired Power Units: A Non-parameter Approach

Energy efficiency analysis has focused predominantly on reducing energy at its point of use by individual components, processes, or products. Efforts to address the energy embodied in materials through strategies and technologies that reduce raw material production - while maintaining the function and use of materials - offer a complementary and expansive approach to traditional energy efficiency analysis. The U.S. Department of Energy (DOE) energy analysis community currently lacks analytical tools that are appropriate for 1) estimating on a product and economy-wide basis energy efficiency potential achieved through material efficiency strategies and 2) evaluating how these new strategies may complement existing energy efficiency efforts. The goal of this report is to describe a standardized analysis framework that would enable calculation of embodied and direct energy and the estimation of energy reduction potential achieved through changes in material use. The proposed framework meets this functionality by spanning the entire U.S. economy, including international trade flows; capturing material and energy use in products and services from raw material extraction to retirement; maintaining material and energy balances; and adjusting dynamically to absolute and relative changes in materials and energy use. The framework is composed of two main elements: a dynamic hybrid input-output (IO) model and a material energy identity. The hybrid IO model defines recipes for material and energy flows in physical (e.g., British thermal units [Btus] and tons) and monetary units, and allows for dynamic adjustments to the amount of materials and energy used. The second element of the analytical framework, the material energy identity, acts as a way of identifying and organizing energy and material efficiency strategies, which are then implemented and analyzed in the hybrid IO model. The report also provides a simple example application of the proposed framework, which illustrates the concept of reducing energy use through changes to material efficiency.

The rapid economy expansion in China has substantially increased energy consumption. Under the stringent environmental policy and the requirement of green economy development, the accurate assessment and analysis of energy efficiency is an increasingly significant issue for energy development policy making in China. This study uses the weighted slacks-based model (weighted SBM) considering the energy substitutability to evaluate the regional energy efficiency (EE) in 29 Chinese provinces, from 1991 to 2015, and explores the sustainable evolution characteristics of EE by comparative and convergence analyses from different perspectives. The empirical results show that EE has significant geographic differences. On the one hand, EE increases from the west to the east of China, and its volatility has a rising trend over the period 1991–2015. Only the EE in the eastern area had a stable rising trend, and the EE differences are difficult to reduce in the short term. On the other hand, the economic zones in the south of China, such as Central Bohai, Pearl River Delta, and Yangtze River Delta, have higher EE. We also find a significant EE improvement occurred during the Eleventh and the Twelfth Five-Year plans. By means of the convergence analysis of energy efficiency across different areas and economic zones over different time intervals, it is shown that EE in the southeast coast provinces have a better catching-up effect and adjustment rate toward the efficient frontier, while the western inland provinces are less effective over the period 1991–2005. Further, we empirically find that the industry policies including industry transfer policy promote EE globally, but the regional differences and fluctuations in EE remain serious. Certain policy implications are discussed with regard to sustainable regional development and an effective industry transfer policy.

As a key equipment in the production process, air compressor is responsible for the most power consumption among industrial equipments. In order to reduce energy consumption of air compressor, its operation condition should be monitored firstly. Although many air compressor monitoring systems can detect operation status, there are still many problems such as complicated installation, inaccurate data and neglected energy efficiency analysis. To this end, the system for air compressor based on wireless data transmission technology was developed, which can realize the function of real-time data display, energy efficiency analysis, historical data viewing and reports export, etc. In order to verify reliability of the system, a field test of oil injected screw air compressor (Atlas G15PA10FM) was carried out in Suzhou Institute of Metrology. The results showed that the detection data not only was accurate and reliable but also can reflect the running condition of air compressor in real time. Through the energy efficiency analysis, it was found that the operating efficiency of the air compressor was less than 20%. Moreover, specific power and electricity consumption per unit exceeds the standard. As a result, the energy waste is serious, which needs improvement for energy-saving. In conclusion, the developed system has a widely applied prospect in the field of air compressor detection.

Cloud computing is a novel paradigm, where the limitations of ubiquitous connected devices in terms of computing, data access, networking and storage are solved through the use of cloud infrastructure. The pervasive adoption of cloud computing results in a rising carbon footprint due to the high energy consumption of computing servers. This negatively affects the environment and entails an associated increase in electricity costs and consequently operational costs. Many works proposed scheduling algorithms using software-centric power models in order to predict electric power consumption in underlying data centers and to schedule cloud tasks so as to reduce energy consumption. Linear models which are based on the lowest-and highest-power data points (referred to here as the “Power Endpoints Model” - PEM) and the simple linear regression (SLR) model are the most used in the literature. However, these models have traditionally been evaluated using different environments, experimental setups, workloads, and error calculation formulas. In this paper, a unified classification and evaluation for these linear power models is presented, under unified setup, benchmarking applications, and error formula with the main goal being to achieve an objective comparison. A new power model is proposed, named Locally Corrected Multiple Linear Regression (LC-MLR), in order to increase prediction accuracy. A simulation framework for a cloud energy-aware scheduler is introduced. The framework combines the Energy-Aware Task Scheduling on Cloud Virtual Machines (EATSVM) with the LC-MLR power model, and facilitates performance measurement for cloud data centers. The scheduler with the new power model increases energy efficiency without degrading the qualities of service of the system. The workloads used for performance evaluation and comparisons in this work are generated using a diverse set of applications. The results show that LC-MLR outperforms the most-used models for simulation of power consumption of cloud data centers. The detailed performance analysis is elaborated in the paper.

Many regular business operations are transforming into digital services, increasing advanced multi-platforms, rapid operational alignment, flexibility, and environmental impact through energy consumption, hardware waste, and technology investments. Flexible and sustainable system development models emphasizing energy efficiency can help innovate software development as digital servicing applications shift. This research is motivated by the need to improve energy consumption in early software design and development due to rising technological efficiency and sustainability demands. Although effective in iterative development and stakeholder engagement, traditional Agile methodologies often struggle with long-term sustainability and energy efficiency. Extended Agile, combining Agile, layered architecture, and aspect-oriented frameworks (ALAI), promises to improve system modularity, flexibility, maintainability, and sustainability. This study’s findings are not just theoretical, but also practically relevant, as they explore the energy efficiency of ALAI software development methodologies, using graduate admission information system services (GAISS) as an example. GAISS is a complex system that handles the entire process of graduate admissions, from application submission to final decision. The study quantifies the energy usage of a student-list webpage by analyzing Microsoft IIS server logs from February 2022 to May 2024. Directly applicable findings show that the GAISS based on the ALAI framework reduces energy consumption by 10.7914% compared to traditional Agile software developments. ALAI used 892.80 kWh versus Agile’s 1000.80 kWh during operations, saving energy. These findings demonstrate the benefits of integrating aspect-oriented frameworks and layering approaches into Agile methodologies, contributing to sustainable software development discourse. The study emphasizes the importance of energy-efficient frameworks such as ALAI to reduce software systems’ environmental impact and promote software development sustainability. The findings of this study, with their practical relevance, assist software developers and organizations in choosing software design and development methods that maximize operational efficiency and environmental sustainability.

According to the characteristics of mechanical equipment remanufacturing system, the dynamic performance of energy consumption and emission is explored, the equipment energy efficiency and emission analysis model is established firstly, and then energy and ecological efficiency analysis method of the remanufacturing system is put forward, at last, the energy and ecological efficiency of WD615.87 automotive diesel engine remanufacturing system as an example is analyzed, the way of energy efficiency improvementnt and environmental friendly mechanism of remanufacturing process is put forward.

Power consumption has become a major design constraint in the embedded systems domain and techniques such as dynamic voltage and frequency scaling (DVFS) have emerged to enhance the system's power and energy efficiency. DVFS-enabling voltage regulators influence the performance, power and energy efficiency of such systems, however, this impact is often neglected or considered late in the design process. In this work, we propose DVFS hardware extensions to a power emulation approach for modeling the voltage regulator behavior, which allows for performance, power and energy efficiency investigations of DVFS-enabled embedded systems. The power emulation approach delivers real-time power information in an early design phase, which allows for the exploration of DVFS efficiency before silicon is available. This offers greater freedom to designers to determine the most apt voltage regulator yielding a system that meets performance, power and energy constraints.

International audience

In this work, we propose a new paradigm called power emulation, which exploits hardware acceleration to drastically speedup power estimation. Power emulation is based on the observation that most power estimation tools typically perform the following sequence of operations: simulating the circuit to obtain value traces or statistics for the inputs of its constituent components, evaluating power models for each circuit component based on the input values seen during simulation, and aggregating the power consumption of individual components to compute the circuit's power consumption. We further recognize that the steps involved in power estimation (power model evaluation, aggregation) can themselves be thought of as synthesizable functions and implemented as hardware circuits. Thus, any given design can be enhanced by adding to it .power estimation hardware., and the resulting power model enhanced circuit can be mapped onto a hardware prototyping platform. While drastic speedups in power estimation (orders of magnitude) are possible using this approach, a significant challenge arises due to the increase in circuit size as a result of adding power estimation hardware. We propose a systematic methodology to reduce the size of the power model enhanced circuit through a suite of techniques, including power model reuse across different circuit components, regulating the granularity of components for power modeling, exploiting inter-component power correlations, resource sharing for power model computations, and the use of block memories for efficient storage within power models. We demonstrate the benefits of the proposed power emulation paradigm by applying it to register-transfer level (RTL) power estimation for industrial designs, resulting in speedups from around 10X to over 500X compared to state-of-the-art commercial power estimation tools.

Skid-Steer Mobile Robots (SSMRs) provide a robust and simple mechanical drive platform making them useful in many applications. Power consumption is an important consideration in the design of any mobile robot and particularly important for SSMRs because of the slipping and corresponding friction that induce large loads on the drive system. The slipping behavior is generally characterized through Instantaneous Centers of Rotation (ICR) of the contact patches and it has been established that these are functions of the system dynamics. However, the existing SSMR power models generally treat these constraints at the kinematic level by assuming constant slip rates taken from empirical data. This paper demonstrates a method to evaluate SSMR power consumption based on slip parameters that are calculated as differential equations extracted from the equations of motion. The dynamic power model is validated and then implemented on two practical manufacturing applications in which a mobile robot is climbing on steel surfaces with primary power consumption due to turning and overcoming gravity. The applications show that the dynamic ICR model plays a significant role in estimating power requirements. The results further demonstrate that the power and energy requirements for a given task depend on the payload and motion along the task in a non-linear fashion, for example showing that the minimum payload does not necessarily correspond to the minimum energy use. This outcome suggests that dynamic effects can be used to find optimal trajectories to minimize power or energy requirements.