Exploratory data analysis for data center energy management

2014

The energy consumption of data centers has become a key issue in today's ICT sector and a significant factor of green environment. A substantial reduction in energy consumption can be made by powering down servers when they are not in use. In cloud data center, it is very hard to manage and allocate their resource to incoming dynamic workload demands. Predicting the required resource demand, it can save the data center's resource wasting and achieve max-profit and min-risk. Without proper prediction, data center can have overprovision and underprovision, which can cause resource waste and significant financial penalties. So an efficient resource management scheme is needed to reduce energy consumption and carbon dioxide (CO2 ) emission. The aim of the present study is to develop model for predicting the future resource demand and estimation of CO2 emission by comparatively assessing the suitability of several machine learning techniques. In order to reduce processing overhea...

International Journal of Electrical and Computer Engineering (IJECE), 2022

Data center hosting is in higher demand to fulfill the computing and storage requirements of information technology (IT) and cloud services platforms which need more electricity to power on the IT devices and for data center cooling requirements. Because of the increased demand for data center facilities, optimizing power usage and ensuring that data center energy quality is not compromised has become a difficult task. As a result, various machine learning-based optimization approaches for enhancing overall power effectiveness have been outlined. This paper aims to identify and analyze the key ongoing research made between 2015 and 2021 to evaluate the types of approaches being used by researchers in data center energy consumption optimization using Machine Learning algorithms. It is discussed how machine learning can be used to optimize data center power. A potential future scope is proposed based on the findings of this review by combining a mixture of bioinspired optimization and neural network.

2016 International Conference on Cloud Computing Research and Innovations (ICCCRI), 2016

Machine learning (ML) approach to modeling and predicting real-world dynamic system behaviours has received widespread research interest. While ML capability in approximating any nonlinear or complex system is promising, it is often a black-box approach, which lacks the physical meanings of the actual system structure and its parameters, as well as their impacts on the system. This paper establishes a model to provide explanation on how system parameters affect its output(s), as such knowledge would lead to potential useful, interesting and novel information. The paper builds on our previous work in ML, and also combines an evolutionary artificial neural networks with sensitivity analysis to extract and validate key factors affecting the cloud data center energy performance. This provides an opportunity for software analysts to design and develop energyaware applications and for Hadoop administrator to optimize the Hadoop infrastructure by having Big Data partitioned in bigger chunks and shortening the time to complete MapReduce jobs.

IAEME PUBLICATION, 2024

This article explores the transformative impact of Artificial Intelligence (AI) and Machine Learning (ML) on energy efficiency in data centers. It examines various areas where these technologies drive significant improvements, including predictive analytics, dynamic cooling management, smart workload scheduling, automated peak shaving, real-time optimization, enhanced maintenance strategies, and holistic system integration. Implementing AI/ML solutions optimizes operational costs and contributes to sustainability efforts by reducing the overall carbon footprint of data center facilities. Through continuous analysis, immediate adjustments, and unified optimization across subsystems, AI-driven approaches achieve substantial energy savings while maintaining or improving performance The article discusses the current state of AI applications in data centers and looks ahead to potential future developments in this rapidly evolving field.

2014

Abstract: Data centers have become an essential operational component of nearly every sector of the economy, and as a result growing consumers of energy and emitters of greenhouse gases (GHGs). Developing strategies for optimizing power usage and reducing the associated life cycle GHG emissions are critical priorities for meeting climate policy objectives. We investigate data center power management through virtualization, a technique that consolidates data center workloads onto fewer computing resources within a data center and deploys computing resources only as needed. Based on an experimentally validated dynamic resource provisioning framework applied to a small scale computing cluster at Drexel University that employed lookahead control, a control scheme using virtualization demonstrated a 25 % reduction in power consumption over a 24-hour period. Using the power savings results from the virtualization experiments, and extrapolating those savings to a medium-sized data center t...

Information

Due to the high demands of new technologies such as social networks, e-commerce and cloud computing, more energy is being consumed in order to store all the data produced and provide the high availability required. Over the years, this increase in energy consumption has brought about a rise in both the environmental impacts and operational costs. Some companies have adopted the concept of a green data center, which is related to electricity consumption and CO2 emissions, according to the utility power source adopted. In Brazil, almost 70% of electrical power is derived from clean electricity generation, whereas in China 65% of generated electricity comes from coal. In addition, the value per kWh in the US is much lower than in other countries surveyed. In the present work, we conducted an integrated evaluation of costs and CO2 emissions of the electrical infrastructure in data centers, considering the different energy sources adopted by each country. We used a multi-layered artifici...

2020

Data centers have become fundamental pillars of the network infrastructures of the various companies or entities regardless of their size. Since they support the processing, analysis, assurance of the data generated in the network, and by the applications in the cloud, which every day increases its volume thanks to diverse and sophisticated technologies. The management and storage of this large volume of information make the data centers consume a lot of energy, generating great concern to owners and administrators. Green Data Center (GDC) is a solution for this problem, reducing the impact produced by the data centers in the environment through the monitoring and control of these and to the application of standards-based on metrics. Although each data center has its particularities and requirements, the metrics are the tools that allow us to measure the energy efficiency of the data center and evaluate if it is friendly to the environment (1.Adv. Intell. Syst. Comput. 574:329-340). The objective of the study is to calculate these metrics in the data centers of a Higher Education Institution in Barranquilla, on both campuses, and the analysis of these will be carried out. It is planned to extend this study by reviewing several metrics to conclude, which is the most efficient and which allows defining the guidelines to update or convert the data center in a friendly environment. The research methodology used for the development of the project is descriptive and noexperimental.

2013

contracted with Syska Hennessy Group to provide technical expertise and resources, including on-site data collection, software expertise, and reporting support. Thanks to Ian Kucma, the lead Syska Hennessy Group engineer for this project and to Vali Sorell for overall technical guidance. Thanks to Romonet for providing software technical support and guidance; specifically Liam Newcombe, Jim Rodriquez, and Gavin Cummings.

ijmer.com

This paper addresses the issue of energy performance of data centers (Server Farms). The primary objective of the study is to understand an empirical energy use pattern of data centers under various conditions like tropical climatic conditions, thermal management problems (HVAC) which consumes about 60 -70 % of the energy, and suggest strategies and give guidance for data centers' design, operation and maintenance and retrofitting to achieve better energy performance. Identification of different zones in a data center, actual energy use characteristics, the airflow and temperature patterns, cooling of electronic equipments, chilled air distribution, lightening system, design criteria, and energy and cost saving potentials were analyzed and compared. Methodology of energy performance evaluation of data centers was discussed. The study concludes that data centers were high energy consuming areas, substantially the HVAC area. Power demands were often grossly over-provided in these facilities. This leaded to substantial increase in capital, running cost, harmful environmental effects. Approximately 40% of energy consumption could be conserved through monitoring, evaluation of implemented conservation measures, efficient designs of base infrastructure and energy consuming systems, practical benchmarking, latest technologies and better practices.

2018 17th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)

Data centers account for approx. 1.4% of the world's electricity consumption, of which up to 50% of that power is dedicated to keeping the actual equipment cool. This represents a huge opportunity to reduce data center energy consumption by tackling the cooling system operations with focus on thermal management. This work presents a novel Data Driven Predictive Model (DDPM) for temperature prediction of server inlet temperatures that utilises high resolution empirical temperature measurements from 52 real-life data centers. A knowledge-base of temperature data and related physical features, created via clustering techniques was used to train a series of artificial neural networks (ANN). The ANNs are used to make predictions of server inlet temperatures based on inputs which describe the boundary conditions. The temperature predictions are made for each server rack to estimate the vertical temperature distribution (s-curve) from the bottom to top of the rack spaced at one foot intervals. Each ANN predicts a temperature at a corresponding vertical height for the given inputs, producing the s-curve reconstructed from the combination of ANN outputs. Furthermore, one ANN predicts the s-curve cluster which is used to provide a prediction confidence. The model only requires local boundary conditions such as rack power, perforated tile airflow rate and temperature, ceiling temperature and a rack adjacency identifier (RAI), in addition to average delta T for air conditioners and an identifier for data center layout type. Both RAI and data center identifier are assigned programmatically given a data centers layout information while the rack power, perforated tile airflow rate and delta T can be measured, metered or calculated. Unlike other statistical approaches which are specific to a single data center room, the DDPM is trained with data from a wide range of data centers, therefore can be used to predict server inlet temperatures in many different types of data centers. Performance of the model on an Intel i7 powered laptop computer with 16GB memory is in the order of 660,000 temperature location predictions per second. The prediction is accompanied by a confidence level in the prediction. DDPM results gave a prediction accuracy of 0.76C RMSE with a 0.12 probability of one point on the s-curve crossing the upper or lower bounds of the confidence interval. The model can perform in real-time, giving way to applications for real-time monitoring, input to optimize control of air conditioning units, and can complement sensor networks.