一个先进的功率和热优化高密度机架解决方案的数据中心能源效率

Jie Zhao, Xianguang Tan, Yajun Pan, Yong Sheng, Hongmei Liu, Jiajun Zhang, Jing Liu, Jun Zhang, Carrie Chen, N. Ahuja
{"title":"一个先进的功率和热优化高密度机架解决方案的数据中心能源效率","authors":"Jie Zhao, Xianguang Tan, Yajun Pan, Yong Sheng, Hongmei Liu, Jiajun Zhang, Jing Liu, Jun Zhang, Carrie Chen, N. Ahuja","doi":"10.1109/ITherm45881.2020.9190394","DOIUrl":null,"url":null,"abstract":"Higher computing capability, higher density and better energy efficiency have always been the goal of data centers. Cross different factors, IT equipment, specifically as server is playing the crucial role in a data center development. CPU/GPU and other hardware ingredients’ computing capability keeps growth with innovated architecture and procedure upgrade, while comes along with increasing server power consumption as well as challenge for energy efficiency. Compared with conventional server system with dedicated fans and power supply units, design optimizations from rack level has been leading evolution to help data center to achieve those goals.This paper introduces an advanced rack scale power and thermal solution pilot deployed in Baidu new self-build data center. With the design, total rack power capacity is up to 36KW, which brings significant improvement for node density. Moreover,48V power distribution is implemented for better rack power efficiency, and centralized power shelf design with dual-input power source halves total PSU quantity for rack space saving and node density increase. Thermal solution is addressed from rack level as well, centralized fan wall provides air flow for whole rack, and thermosyphon heatsink is implemented in each node from higher heat dissipate efficiency perspective. With all the advanced technology integrated and optimized together, detailed energy efficiency contribution to Total Cost Ownership (TCO) is summarized and compared with conventional rack design and shows the advantages for data center energy efficiency.","PeriodicalId":193052,"journal":{"name":"2020 19th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"An Advanced Power and Thermal Optimized High Density Rack Solution for Data Center Energy Efficiency\",\"authors\":\"Jie Zhao, Xianguang Tan, Yajun Pan, Yong Sheng, Hongmei Liu, Jiajun Zhang, Jing Liu, Jun Zhang, Carrie Chen, N. Ahuja\",\"doi\":\"10.1109/ITherm45881.2020.9190394\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Higher computing capability, higher density and better energy efficiency have always been the goal of data centers. Cross different factors, IT equipment, specifically as server is playing the crucial role in a data center development. CPU/GPU and other hardware ingredients’ computing capability keeps growth with innovated architecture and procedure upgrade, while comes along with increasing server power consumption as well as challenge for energy efficiency. Compared with conventional server system with dedicated fans and power supply units, design optimizations from rack level has been leading evolution to help data center to achieve those goals.This paper introduces an advanced rack scale power and thermal solution pilot deployed in Baidu new self-build data center. With the design, total rack power capacity is up to 36KW, which brings significant improvement for node density. Moreover,48V power distribution is implemented for better rack power efficiency, and centralized power shelf design with dual-input power source halves total PSU quantity for rack space saving and node density increase. Thermal solution is addressed from rack level as well, centralized fan wall provides air flow for whole rack, and thermosyphon heatsink is implemented in each node from higher heat dissipate efficiency perspective. With all the advanced technology integrated and optimized together, detailed energy efficiency contribution to Total Cost Ownership (TCO) is summarized and compared with conventional rack design and shows the advantages for data center energy efficiency.\",\"PeriodicalId\":193052,\"journal\":{\"name\":\"2020 19th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)\",\"volume\":\"13 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2020-07-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2020 19th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ITherm45881.2020.9190394\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2020 19th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ITherm45881.2020.9190394","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2

摘要

更高的计算能力、更高的密度和更高的能源效率一直是数据中心的目标。在不同的因素中,IT设备,特别是服务器在数据中心的发展中起着至关重要的作用。CPU/GPU等硬件组件的计算能力随着架构的创新和过程的升级而不断增长,同时也伴随着服务器功耗的增加和能效的挑战。与具有专用风扇和电源单元的传统服务器系统相比,机架级的设计优化一直在引领发展,以帮助数据中心实现这些目标。介绍了在百度新建自建数据中心部署的先进机架级电源和热解决方案试点。设计后,机架总功率可达36KW,大大提高了节点密度。采用48V配电,提高了机架的电源效率;采用双输入电源的集中式电源机箱设计,使电源模块总数减半,节省了机架空间,提高了节点密度。从机架层面解决散热问题,集中式风扇墙为整个机架提供气流,各节点采用热虹吸散热器,提高散热效率。通过对所有先进技术的集成和优化,总结了详细的能源效率对总成本拥有(TCO)的贡献,并与传统机架设计进行了比较,展示了数据中心能源效率的优势。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
An Advanced Power and Thermal Optimized High Density Rack Solution for Data Center Energy Efficiency
Higher computing capability, higher density and better energy efficiency have always been the goal of data centers. Cross different factors, IT equipment, specifically as server is playing the crucial role in a data center development. CPU/GPU and other hardware ingredients’ computing capability keeps growth with innovated architecture and procedure upgrade, while comes along with increasing server power consumption as well as challenge for energy efficiency. Compared with conventional server system with dedicated fans and power supply units, design optimizations from rack level has been leading evolution to help data center to achieve those goals.This paper introduces an advanced rack scale power and thermal solution pilot deployed in Baidu new self-build data center. With the design, total rack power capacity is up to 36KW, which brings significant improvement for node density. Moreover,48V power distribution is implemented for better rack power efficiency, and centralized power shelf design with dual-input power source halves total PSU quantity for rack space saving and node density increase. Thermal solution is addressed from rack level as well, centralized fan wall provides air flow for whole rack, and thermosyphon heatsink is implemented in each node from higher heat dissipate efficiency perspective. With all the advanced technology integrated and optimized together, detailed energy efficiency contribution to Total Cost Ownership (TCO) is summarized and compared with conventional rack design and shows the advantages for data center energy efficiency.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信