Font Size: a A A

Design And Implementation Of Energy Management System For Data Center

Posted on:2018-02-19Degree:MasterType:Thesis
Country:ChinaCandidate:L Q YangFull Text:PDF
GTID:2428330590492439Subject:Software engineering
Abstract/Summary:PDF Full Text Request
With the concept of Internet + big data and cloud business presents the geometric form of growth,the demand for data frame mounting.In 2016,according to domestic operators has reached 63.6 billion yuan,the electricity expenses province company in the field of electricity energy consumption spending alone reached 7.6 billion yuan.Single stand power consumption is also rising,data center energy consumption have been unable to compared with the traditional energy consumption.Data center energy consumption needs to build an accurate detection,measurement and analysis system,applicable to the data center power consumption management system become daily operations focus on the problem.Based on operators geographically distributed data centers,construction equipment complex,large difference of levels of energy consumption of data centres,multiple data acquisition methods,no unified data analysis model of energy consumption,etc.,based on the data center energy consumption of the actual operation requirements and improve the capacity management system set up project,studied the application in the complex data center scenario,energy management system design and implementation process.Two aspects of the system is mainly to solve the demand,the energy consumption statistics show,the second is energy consumption data analysis,the paper work is mainly manifested in the following two aspects:1)USES the object-oriented method,the demand of the system are analyzed in detail the business scenario.Propose solutions to realize data center electricity energy consumption structure,PUE parameter of unified management,but also provides electricity analysis,the different equipment energy consumption etc,to provide users with the base of the data center operation data.2)in terms of energy consumption data analysis,energy consumption analysis model based on the data center,with the aid of DBSCAN clustering methods,such as C4.5 decision tree method,implements the comprehensive energy consumption data query,the energy consumption structure analysis and evaluation of energy saving features,analysis of energy consumption model focuses on the optimization for the daily operation,used for data center daily energy consumption of anomaly detection and evaluation of data center energy consumption level.3)application of Flex technology build a platform to show,make the energy consumption data is clear;Based on the.net implements a high reusability and strong extensibility architecture;At the same time to verify the clustering recognition for the reliability of the data center energy consumption model judgment.System in the trial run on system design application scenarios,and the main function was tested,from the system is suitable for data center operations scenario,meet the design requirements.Through the mode analysis was carried out on the acquisition of energy consumption data,set up under different construction scheme,equipment configuration standards for energy consumption model of the data center.For the new data center in the location conditions,construction scheme selection,energy-saving technology selection,equipment selection,operation mode,etc.,to provide data analysis support.At the end of the paper is aiming at the condition of the system appeared in the process of application,the system has made the reasonable planning and prospect.
Keywords/Search Tags:Energy consumption analysis model, Energy management, PUE
PDF Full Text Request
Related items