scholarly journals Towards an Autonomous Framework for HPC Optimization: Using Machine Learning for Energy and Performance Modeling

2019 ◽  
Author(s):  
Vinícius Klôh ◽  
Matheus Gritz ◽  
Bruno Schulze ◽  
Mariza Ferro
Keyword(s):  
Machine Learning ◽  
Energy Efficiency ◽  
Energy Consumption ◽  
Energy Saving ◽  
High Performance ◽  
Performance Modeling ◽  
Application Performance ◽  
Ongoing Research ◽  
And Performance

Performance and energy efficiency are now critical concerns in high performance scientific computing. It is expected that requirements of the scientific problem should guide the orchestration of different techniques of energy saving, in order to improve the balance between energy consumption and application performance. To enable this balance, we propose the development of an autonomous framework to make this orchestration and present the ongoing research to this development, more specifically, focusing in the characterization of the scientific applications and the performance modeling tasks using Machine Learning.

scholarly journals HEMS-IoT: A Big Data and Machine Learning-Based Smart Home System for Energy Saving

Energies ◽  
2020 ◽  
Vol 13 (5) ◽  
pp. 1097 ◽  
Author(s):  
Isaac Machorro-Cano ◽  
Giner Alor-Hernández ◽  
Mario Andrés Paredes-Valverde ◽  
Lisbeth Rodríguez-Mazahua ◽  
José Luis Sánchez-Cervantes ◽  
...  
Keyword(s):  
Machine Learning ◽  
Energy Efficiency ◽  
Big Data ◽  
Energy Consumption ◽  
Energy Saving ◽  
Smart Home ◽  
User Preferences ◽  
The Internet ◽  
Automation Systems ◽  
Home Energy Management

Energy efficiency has aroused great interest in research worldwide, because energy consumption has increased in recent years, especially in the residential sector. The advances in energy conversion, along with new forms of communication, and information technologies have paved the way for what is now known as smart homes. The Internet of Things (IoT) is the convergence of various heterogeneous technologies from different application domains that are used to interconnect things through the Internet, thus allowing for the detection, monitoring, and remote control of multiple devices. Home automation systems (HAS) combined with IoT, big data technologies, and machine learning are alternatives that promise to contribute to greater energy efficiency. This work presents HEMS-IoT, a big data and machine learning-based smart home energy management system for home comfort, safety, and energy saving. We used the J48 machine learning algorithm and Weka API to learn user behaviors and energy consumption patterns and classify houses with respect to energy consumption. Likewise, we relied on RuleML and Apache Mahout to generate energy-saving recommendations based on user preferences to preserve smart home comfort and safety. To validate our system, we present a case study where we monitor a smart home to ensure comfort and safety and reduce energy consumption.

ENERGY SAVING TECHNOLOGIES IN POWER SUPPLY OF SMART BUILDINGS USING ARTIFICIAL INTELLIGENCE

2020 ◽  
pp. 29-37
Author(s):  
E. Yu. Shchetinin
Keyword(s):  
Machine Learning ◽  
Energy Efficiency ◽  
Energy Consumption ◽  
Energy Saving ◽  
Commercial Buildings ◽  
Machine Learning Algorithms ◽  
Gradient Boosting ◽  
Digital Measurement ◽  
Computer Algorithms ◽  
Smart Buildings

Intelligent energy saving and energy efficiency technologies are the modern large-scale global trend in the energy systems development. The demand for smart buildings is growing not only in the world, but also in Russia, especially in the market of construction and operation of large business centers, shopping centers and other business projects. Accurate cost estimates are important for promoting energy efficiency construction projects and demonstrating their economic attractiveness. The growing number of digital measurement infrastructure, used in commercial buildings, led to increase access to high-frequency data that can be used for anomaly detection and diagnostics of equipment, heating, ventilation, and optimization of air conditioning. This led to the use of modern and efficient machine learning methods that provide promising opportunities to obtain more accurate forecasts of energy consumption of the buildings, and thus increase energy efficiency. In this paper, based on the gradient boosting model, a method of modeling and forecasting the energy consumption of buildings is proposed and computer algorithms are developed to implement it. Energy consumption dataset of 300 commercial buildings was used to assess the effectiveness of the proposed algorithms. Computer simulations showed that the use of these algorithms has increased the accuracy of the prediction of energy consumptionin more than 80 percent of cases compared to other machine learning algorithms.

High Performance Buildings: The Impact of Building Orientation on HVAC System Loads

2017 ◽  
Author(s):  
Rushang B. Shah ◽  
Navid Goudarzi
Keyword(s):  
Energy Efficiency ◽  
Renewable Energy ◽  
Energy Consumption ◽  
Energy Saving ◽  
Energy Demand ◽  
High Performance ◽  
Cost Effective ◽  
Building Energy ◽  
Hvac System ◽  
Building Orientation

Energy demand growth and depletion of conventional energy resources in recent years have led to exploring alternative energy resources and further concentration on improving energy efficiency of segments with higher energy consumption. Building energy demand is among the main areas of concern with a 40% average of total energy consumption in the US energy market. Within building energy demand approximately, the Heating, Ventilation and Air Conditioning (HVAC) system, lighting system, has the largest energy consumption share compared to other systems such as electronics systems, water-heating-cooking, and other systems. This implies that small improvements in HVAC system loads will result in significant energy savings. Novel cost-effective solutions should be developed to integrate and optimize all the essential high-performance building attributes, especially energy efficiency and occupant productivity. Employing comprehensive building energy analysis (BEA) simulation tools are among the cheapest, yet are the cost-effective approaches in improving building energy performance. This paper follows the energy saving practice using existing BEA simulation tools with a focus on two major aspects that can contribute to building thermal loads: building orientation and integrating renewable energy. The results show the significant impact of building orientation for developing energy efficiency solutions with focus on integrating renewable energy technologies within high performance buildings. This work provides a basis for the follow on phases of this research to develop smart energy saving solutions using current BEA simulation platforms. Such adds-on features enable users to improve building energy saving by determining building design features and integrating renewable energy solutions based on identified optimal building orientations.

The Research of Measuring Approach and Energy Efficiency for Hadoop Periodic Jobs

2015 ◽  
Vol 8 (1) ◽  
pp. 206-210 ◽  
Author(s):  
Yu Junyang ◽  
Hu Zhigang ◽  
Han Yuanyuan
Keyword(s):  
Energy Efficiency ◽  
Cloud Computing ◽  
Energy Consumption ◽  
Energy Saving ◽  
Energy Estimate ◽  
Cloud Platform ◽  
Periodic Tasks ◽  
Current Consumption ◽  
Reduce Energy Consumption

Current consumption of cloud computing has attracted more and more attention of scholars. The research on Hadoop as a cloud platform and its energy consumption has also received considerable attention from scholars. This paper presents a method to measure the energy consumption of jobs that run on Hadoop, and this method is used to measure the effectiveness of the implementation of periodic tasks on the platform of Hadoop. Combining with the current mainstream of energy estimate formula to conduct further analysis, this paper has reached a conclusion as how to reduce energy consumption of Hadoop by adjusting the split size or using appropriate size of workers (servers). Finally, experiments show the effectiveness of these methods as being energy-saving strategies and verify the feasibility of the methods for the measurement of periodic tasks at the same time.

scholarly journals Statistical and machine learning models for optimizing energy in parallel applications

2019 ◽  
Vol 33 (6) ◽  
pp. 1079-1097 ◽  
Author(s):  
Mark Endrei ◽  
Chao Jin ◽  
Minh Ngoc Dinh ◽  
David Abramson ◽  
Heidi Poxon ◽  
...  
Keyword(s):  
Machine Learning ◽  
Energy Efficiency ◽  
High Performance ◽  
Large Scale ◽  
Energy Use ◽  
Parallel Applications ◽  
Learning Models ◽  
Trade Off ◽  
Time Required ◽  
Machine Learning Models

Rising power costs and constraints are driving a growing focus on the energy efficiency of high performance computing systems. The unique characteristics of a particular system and workload and their effect on performance and energy efficiency are typically difficult for application users to assess and to control. Settings for optimum performance and energy efficiency can also diverge, so we need to identify trade-off options that guide a suitable balance between energy use and performance. We present statistical and machine learning models that only require a small number of runs to make accurate Pareto-optimal trade-off predictions using parameters that users can control. We study model training and validation using several parallel kernels and more complex workloads, including Algebraic Multigrid (AMG), Large-scale Atomic Molecular Massively Parallel Simulator, and Livermore Unstructured Lagrangian Explicit Shock Hydrodynamics. We demonstrate that we can train the models using as few as 12 runs, with prediction error of less than 10%. Our AMG results identify trade-off options that provide up to 45% improvement in energy efficiency for around 10% performance loss. We reduce the sample measurement time required for AMG by 90%, from 13 h to 74 min.

scholarly journals Power and Performance Evaluation of Memory-Intensive Applications

Energies ◽  
2021 ◽  
Vol 14 (14) ◽  
pp. 4089
Author(s):  
Kaiqiang Zhang ◽  
Dongyang Ou ◽  
Congfeng Jiang ◽  
Yeliang Qiu ◽  
Longchuan Yan
Keyword(s):  
Energy Efficiency ◽  
Energy Consumption ◽  
Power Consumption ◽  
Job Scheduling ◽  
Memory System ◽  
Processor Core ◽  
Memory Efficiency ◽  
And Performance ◽  
Reasonable Use ◽  
Server System

In terms of power and energy consumption, DRAMs play a key role in a modern server system as well as processors. Although power-aware scheduling is based on the proportion of energy between DRAM and other components, when running memory-intensive applications, the energy consumption of the whole server system will be significantly affected by the non-energy proportion of DRAM. Furthermore, modern servers usually use NUMA architecture to replace the original SMP architecture to increase its memory bandwidth. It is of great significance to study the energy efficiency of these two different memory architectures. Therefore, in order to explore the power consumption characteristics of servers under memory-intensive workload, this paper evaluates the power consumption and performance of memory-intensive applications in different generations of real rack servers. Through analysis, we find that: (1) Workload intensity and concurrent execution threads affects server power consumption, but a fully utilized memory system may not necessarily bring good energy efficiency indicators. (2) Even if the memory system is not fully utilized, the memory capacity of each processor core has a significant impact on application performance and server power consumption. (3) When running memory-intensive applications, memory utilization is not always a good indicator of server power consumption. (4) The reasonable use of the NUMA architecture will improve the memory energy efficiency significantly. The experimental results show that reasonable use of NUMA architecture can improve memory efficiency by 16% compared with SMP architecture, while unreasonable use of NUMA architecture reduces memory efficiency by 13%. The findings we present in this paper provide useful insights and guidance for system designers and data center operators to help them in energy-efficiency-aware job scheduling and energy conservation.

scholarly journals Energy Saving Quantitative Analysis of Passive, Active, and Renewable Technologies in Different Climate Zones

2021 ◽  
Vol 11 (15) ◽  
pp. 7115
Author(s):  
Chul-Ho Kim ◽  
Min-Kyeong Park ◽  
Won-Hee Kang
Keyword(s):  
Renewable Energy ◽  
Energy Consumption ◽  
Energy Saving ◽  
High Performance ◽  
Energy Systems ◽  
Outdoor Air ◽  
Climate Zones ◽  
Renewable Energy Systems ◽  
Active Systems ◽  
High Performance Buildings

The purpose of this study was to provide a guideline for the selection of technologies suitable for ASHRAE international climate zones when designing high-performance buildings. In this study, high-performance technologies were grouped as passive, active, and renewable energy systems. Energy saving technologies comprising 15 cases were categorized into passive, active, and renewable energy systems. EnergyPlus v9.5.0 was used to analyze the contribution of each technology in reducing the primary energy consumption. The energy consumption of each system was analyzed in different climates (Incheon, New Delhi, Minneapolis, Berlin), and the detailed contributions to saving energy were evaluated. Even when the same technology is applied, the energy saving rate differs according to the climatic characteristics. Shading systems are passive systems that are more effective in hot regions. In addition, the variable air volume (VAV) system, combined VAV–energy recovery ventilation (ERV), and combined VAV–underfloor air distribution (UFAD) are active systems that can convert hot and humid outdoor temperatures to create comfortable indoor environments. In cold and cool regions, passive systems that prevent heat loss, such as high-R insulation walls and windows, are effective. Active systems that utilize outdoor air or ventilation include the combined VAV-economizer, the active chilled beam with dedicated outdoor air system (DOAS), and the combined VAV-ERV. For renewable energy systems, the ground source heat pump (GSHP) is more effective. Selecting energy saving technologies that are suitable for the surrounding environment, and selecting design strategies that are appropriate for a given climate, are very important for the design of high-performance buildings globally.

scholarly journals Specific Electrical Energy Consumption and CO2 Emissions Assessment of Agrifood Industries in the Central Region of Portugal

2014 ◽  
Vol 675-677 ◽  
pp. 1880-1886 ◽  
Author(s):  
Pedro D. Silva ◽  
Pedro Dinis Gaspar ◽  
J. Nunes ◽  
L.P.A Andrade
Keyword(s):  
Energy Efficiency ◽  
Energy Consumption ◽  
Food Safety ◽  
Central Region ◽  
Electrical Energy ◽  
Electrical Energy Consumption ◽  
Reference Levels ◽  
Energy Characteristics ◽  
Result Analysis

This paper provides a characterization of the electrical energy consumption of agrifood industries located in the central region of Portugal that use refrigeration systems to ensure the food safety. The study is based on the result analysis of survey data and energy characteristics of the participating companies belonging to the following agrifood sectors: meat, dairy, horticultural, distribution and wine. Through the quantification of energy consumption of companies is possible to determine the amount of greenhouse gases (GHGs) emissions indexed to its manufacturing process. Comparing the energy and GHGs emissions indexes of companies of a sector and between sectors is possible to create reference levels. With the results of this work is possible to rating the companies in relation to reference levels of energy and GHGs emissions and thus promote the rational use of energy by the application of practice measures for the improvement of the energy efficiency and the reduction of GHGs emissions.

An Interpretable Machine Learning Model Enhanced Integrated CPU-GPU DVFS Governor

2021 ◽  
Vol 20 (6) ◽  
pp. 1-28
Author(s):  
Jurn-Gyu Park ◽  
Nikil Dutt ◽  
Sung-Soo Lim
Keyword(s):  
Machine Learning ◽  
Energy Efficiency ◽  
High Performance ◽  
Linear Models ◽  
Piecewise Linear ◽  
Prediction Errors ◽  
Linear Regression Models ◽  
Mobile Games ◽  
Interpretable Machine Learning ◽  
Mathematical Formulas

Modern heterogeneous CPU-GPU-based mobile architectures, which execute intensive mobile gaming/graphics applications, use software governors to achieve high performance with energy-efficiency. However, existing governors typically utilize simple statistical or heuristic models, assuming linear relationships using a small unbalanced dataset of mobile games; and the limitations result in high prediction errors for dynamic and diverse gaming workloads on heterogeneous platforms. To overcome these limitations, we propose an interpretable machine learning (ML) model enhanced integrated CPU-GPU governor: (1) It builds tree-based piecewise linear models (i.e., model trees) offline considering both high accuracy (low error) and interpretable ML models based on mathematical formulas using a simulatability operation counts quantitative metric. And then (2) it deploys the selected models for online estimation into an integrated CPU-GPU Dynamic Voltage Frequency Scaling governor. Our experiments on a test set of 20 mobile games exhibiting diverse characteristics show that our governor achieved significant energy efficiency gains of over 10% (up to 38%) improvements on average in energy-per-frame with a surprising-but-modest 3% improvement in Frames-per-Second performance, compared to a typical state-of-the-art governor that employs simple linear regression models.

Sign in / Sign up

Export Citation Format

Share Document