Yuehlin Tsai, Guilin Wang, Yuehong Jin, Sheng Li, Wenyi Fang, Guofeng Chen, Peng Cao, Jun Zhang, Yuehong Fan, Yuyang Xia, Wenbin Tian, Hang Cheng, Xingping Ruan, Hongxing Zhou, N. Ahuja, Qing Qiao, Mohan J. Kumar
{"title":"An Advanced Cold Plate Liquid Cooling Rack Design for Hyperscale Data Center","authors":"Yuehlin Tsai, Guilin Wang, Yuehong Jin, Sheng Li, Wenyi Fang, Guofeng Chen, Peng Cao, Jun Zhang, Yuehong Fan, Yuyang Xia, Wenbin Tian, Hang Cheng, Xingping Ruan, Hongxing Zhou, N. Ahuja, Qing Qiao, Mohan J. Kumar","doi":"10.1109/iTherm54085.2022.9899555","DOIUrl":null,"url":null,"abstract":"As the emerging business models and advanced customer experience is driving an explosion of data at the era of 5G, data center has been evolving with increasingly powerful AI (Artificial Intelligence) computing capability for enhancing CSP (Cloud Service Provider) cloud service by data analytics in data centric world. High-performance computing, storage and networking are essential to achieve this. Meanwhile, energy efficiency, which is measured by PUE, is the mandatory design target for modern data center. The cold plate liquid cooling solution is one of crucial technologies to continual address the demand of higher performance computing at lower PUE. This paper introduces an advanced cold plate liquid cooling rack architecture for JD.COM data center as well as the key learnings from JD.com engineering practice and validation data, including system level liquid cooling design related ingredients’ design and selection, such as CPU cold plates, memory cold plates, leakage detection, and quick disconnects, etc. Moreover, JD.COM’s novel methodology for optimized CPU cold plate design as well as the implementation of whole-loop leakage detection in a JD.COM datacenter are also introduced.","PeriodicalId":351706,"journal":{"name":"2022 21st IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (iTherm)","volume":"80 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-05-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","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/iTherm54085.2022.9899555","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3
Abstract
As the emerging business models and advanced customer experience is driving an explosion of data at the era of 5G, data center has been evolving with increasingly powerful AI (Artificial Intelligence) computing capability for enhancing CSP (Cloud Service Provider) cloud service by data analytics in data centric world. High-performance computing, storage and networking are essential to achieve this. Meanwhile, energy efficiency, which is measured by PUE, is the mandatory design target for modern data center. The cold plate liquid cooling solution is one of crucial technologies to continual address the demand of higher performance computing at lower PUE. This paper introduces an advanced cold plate liquid cooling rack architecture for JD.COM data center as well as the key learnings from JD.com engineering practice and validation data, including system level liquid cooling design related ingredients’ design and selection, such as CPU cold plates, memory cold plates, leakage detection, and quick disconnects, etc. Moreover, JD.COM’s novel methodology for optimized CPU cold plate design as well as the implementation of whole-loop leakage detection in a JD.COM datacenter are also introduced.