针对数据中心机柜级冷却系统存在的缺点，提出行级备份式容灾方案，与传统房间级备份式容灾方案进行对比分析，以单机柜功率、故障空调位置和备份空调位置为研究对象，利用计算流体力学（CFD）模拟软件对不同方案下机柜内设备冷却效果进行模拟。结果表明：采用行级备份式容灾方案的设备冷却效果显著优于房间级精密空调的容灾方案，且能够保证设备工作在正常的温度范围内；行级备份式方案能够保证故障空调附近机柜的进出风温度与周围机柜相同，但房间备份式会出现故障空调附近局部高温，存在热点。
A row-level backup disaster recovery scheme is proposed in this paper for solving the defects of rack-level cooling system in data centers. The cooling performance is compared with traditional room-level backup disaster tolerant scheme using computer fluid dynamics (CFD) simulation software. The simulation results show that the cooling effect of IT equipment under row-level cooling backup disaster tolerant scheme is significantly better than that of room-level backup scheme. The devices can work in normal temperature range when one air conditioner fails in row-level backup scheme, while hot spots will occur in row level backup scheme.
[1] 陈杰. 数据机房冷通道封闭技术应用及模拟分析[J]. 暖通空调, 2015, 45(6): 37-40.
[2] 杨国荣, 胡仰耆, 马伟骏. 数据中心空调设计初探[J]. 建筑电气, 2009, 28(12): 21-26. DOI: 10.3969/j.issn. 1003-8493.2009.12.006.
[3] 中国数据中心产业发展联盟. 2012年中国数据中心产业发展白皮书[R/OL]. 中国数据中心产业发展联盟, 2012. https://wenku.baidu.com/view/c484ced4240c844769eaeeec.html
[4] CHO J, LIM T, KIM B S. Measurements and predictions of the air distribution systems in high compute density (Internet) data centers[J]. Energy and buildings, 2009, 41(10): 1107-1115. DOI: 10.1016/j.enbuild.2009.05.017.
[5] CHO J, KIM B S. Evaluation of air management system's thermal performance for superior cooling efficiency in high-density data centers[J]. Energy and buildings, 2011, 43(9): 2145-2155. DOI: 10.1016/j.enbuild. 2011.04.025.
[6] PATANKAR S V. Airflow and cooling in a data center[J]. Journal of heat transfer, 2010, 132(7): 073001. DOI: 10.1115/1.4000703.
[7] 蒋雅靖, 刘成. 列间空调在高密度数据中心应用的数值模拟[J]. 建筑热能通风空调, 2012, 31(6): 92-95. DOI: 10.3969/j.issn.1003-0344.2012.06.027.
[8] 刘芳, 王志刚. 某数据中心室内空调气流组织的模拟研究[J]. 暖通与空调, 2016, 44(1): 11-17. DOI: 10.3969/j.issn.1673-7237.2016.10.004.
[9] 舒庆鑫. 变电站数据机房能耗及气流组织模拟研究[D]. 杭州: 浙江大学, 2014.
[10] 胡坤, 李振北. ANSYS ICEM CFD工程实例详解[M]. 北京: 人民邮电出版社, 2014: 207-215.
[11] CHEN Q. Comparison of different k-ε models for indoor air flow computations[J]. Numerical heat transfer, part b: fundamentals, 1995, 28(3): 353-369. DOI: 10.1080/ 10407799508928838.
[12] YAKHOT V, ORSZAG S A, THANGAM S, et al. Development of turbulence models for shear flows by a double expansion technique[J]. Physics of fluids a: fluid dynamics, 1992, 4(7): 1510-1520. DOI: 10.1063/1.858424.
[13] KIM S W, CHEN C P. A multiple-time-scale turbulence model based on variable partitioning of the turbulent kinetic energy spectrum[J]. Numerical heat transfer, part b: fundamentals, 1990, 16(2): 193-211. DOI: 10407798908944935.
