{"title":"IT Equipment Cooling Assessement and Metrics","authors":"J. Vangilder, W. Tian, M. Condor","doi":"10.1109/iTherm54085.2022.9899612","DOIUrl":null,"url":null,"abstract":"The primary goal of data center cooling is to ensure that the temperature of the airflow supplied to the inlets of IT equipment does not exceed a threshold specified by industry guidelines or IT vendors. We stress that the critical temperature is the inlet of the IT equipment (inside the rack) and not inlet (i.e., front door) of the rack itself. However, because it is the airflow patterns inside and outside the rack that dictate the temperature of the air that ultimately reaches IT inlets, IT inlet temperature alone is an insufficient metric for optimizing cooling and identifying the root cause of issues like hot spots. We propose that data center designers and operators performing CFD simulation focus on the key airflow patterns inside and outside the rack – which can be quantified by existing capture index metrics. We discuss the key airflow patterns and metrics and provide an example of how they can be used in practice. Finally, as rack inlet and IT inlet temperatures are sometimes (incorrectly) used interchangeably by the data center community, we quantify the difference between the two through the CFD analysis of a real data center; differences are found to be significant even under conditions which tend to minimize internal-rack recirculation.","PeriodicalId":351706,"journal":{"name":"2022 21st IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (iTherm)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-05-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 21st IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (iTherm)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/iTherm54085.2022.9899612","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
The primary goal of data center cooling is to ensure that the temperature of the airflow supplied to the inlets of IT equipment does not exceed a threshold specified by industry guidelines or IT vendors. We stress that the critical temperature is the inlet of the IT equipment (inside the rack) and not inlet (i.e., front door) of the rack itself. However, because it is the airflow patterns inside and outside the rack that dictate the temperature of the air that ultimately reaches IT inlets, IT inlet temperature alone is an insufficient metric for optimizing cooling and identifying the root cause of issues like hot spots. We propose that data center designers and operators performing CFD simulation focus on the key airflow patterns inside and outside the rack – which can be quantified by existing capture index metrics. We discuss the key airflow patterns and metrics and provide an example of how they can be used in practice. Finally, as rack inlet and IT inlet temperatures are sometimes (incorrectly) used interchangeably by the data center community, we quantify the difference between the two through the CFD analysis of a real data center; differences are found to be significant even under conditions which tend to minimize internal-rack recirculation.