Thermal Management of Electronic Servers under Different Power Conditions

An experimental study of thermal management of rack of servers of a data centers is presented for different schemes of server’s power loadings. Four scheme of server’s power loading (uniform, discrete, segmented and clustered) are considered. A total of eleven cases of servers loading conditions are investigated. Temperature distributions and dimensionless measuring indices (SHI/RHI) are used to measure and evaluate the thermal management of the data center. The results showed that (i) uniform power loadingprovides the best results for (SHI/RHI) among the other schemesconsidered where the optimum benefit of cold air in server cooling is obtained at uniformly power loading of servers, (ii) clustering of active servers lead to better air flow management compared to discretely individual active servers and segmented distributions of active servers, and (iii) servers located at the bottom rack cabinet always has better thermal performance compared to servers at higher levels.

[1]  Cullen E. Bash,et al.  DIMENSIONLESS PARAMETERS FOR EVALUATION OF THERMAL DESIGN AND PERFORMANCE OF LARGE-SCALE DATA CENTERS , 2002 .

[2]  C. Belady,et al.  Gravity-assisted air mixing in data centers and how it affects the rack cooling effectiveness , 2006, Thermal and Thermomechanical Proceedings 10th Intersociety Conference on Phenomena in Electronics Systems, 2006. ITHERM 2006..

[3]  Jun Yang,et al.  Comparison of overhead and underfloor air delivery systems in a data center environment usinq CFD modeling , 2005 .

[4]  Roger R. Schmidt,et al.  Airflow Uniformity Through Perforated Tiles in a Raised-Floor Data Center , 2005 .

[5]  Bahgat Sammakia,et al.  Comparative Analysis of Different Data Center Airflow Management Configurations , 2005 .

[6]  Bahram Moshfegh,et al.  Investigation of indoor climate and power usage in a data center , 2005 .

[7]  Hazim B. Awbi,et al.  Scale effect in room airflow studies , 1990 .

[8]  John F. Dannenhoffer,et al.  Design of Simulated Server Racks for Data Center Research , 2011 .

[9]  Jinkyun Cho,et al.  Measurements and predictions of the air distribution systems in high compute density (Internet) data centers , 2009 .

[10]  Yogendra Joshi,et al.  Effect of server load variation on rack air flow distribution in a raised floor data center , 2011, 2011 27th Annual IEEE Semiconductor Thermal Measurement and Management Symposium.

[11]  M.K. Herrlin.  A New Tool for Evaluating and Designing the Thermal Environment in Telecom Central Offices , 2006, INTELEC 06 - Twenty-Eighth International Telecommunications Energy Conference.