No abstract

Complete immersion of servers in synthetic dielectric fluids is rapidly becoming a popular technique to minimize the energy consumed by data centers for cooling purposes. In general, immersion cooling offers noteworthy advantages over conventional air-cooling methods as synthetic dielectric fluids have high heat dissipation capacities which are roughly about 1200 times greater than air. Other advantages of dielectric fluid immersion cooling include even thermal profile on chips, reduction in noise and addressing reliability and operational enhancements like whisker formation and electrochemical migration. Nevertheless, lack of data published and availability of long-term reliability data on immersion cooling is insufficient which makes most of data centers operators reluctant to implement this technique. The first part of this paper will compare thermal performance of single-phase oil immersion cooled HP ProLiant DL160 G6 server against air cooled server using computational fluid dynamics on 6SigmaET®. Focus of the study are major components of the server like Central Processing Unit (CPU), Dual in Line Memory Module (DIMM), Input/output Hub (IOH) chip and Input/output controller Hub (ICH). The second part of this paper focuses on thermal performance optimization of oil immersion cooled servers by varying inlet oil temperature, flow rate and using different fluid.

In a data center, electronic equipment such as server and switches dissipate heat and the corresponding cooling systems contribute to typically 25–35% of total energy consumption. The heat load continues to increase as there is a greater need for miniaturization and convergence. In 2014, data centers in the U.S. consumed an estimated 70 billion kWh, representing about 1.8% of total U.S. electricity consumption. Based on current trend estimates, U.S. data centers are projected to consume approximately 73 billion kWh in 2020 [1]. Many research and strategies are adopted to minimize energy cost. The recommended dry bulb temperature for long-term operation and reliability for air cooling is between 18–27°C and the largest allowable inlet temperature range to operate at is between 5°C and 45°C with American Society of Heating, Refrigeration, and Air-Conditioning Engineers (ASHRAE) enabling much broader allowable zones) [2]. But understanding a proper cooling system is very important especially for thermal management of IT equipment with high heat loads such as 1U or 2U multi-core, high-end servers and blade servers which provide more computing per watt. Many problems like high inlet temperature due to the mixing of hot air with cold air, local hot spots, lower system reliability, increased failure, and downtime may occur. Among many other approaches to managing high-density racks, in-row coolers are used in between racks to provide cold air and minimize local hot spots. This paper describes a computational study being performed by applying in-row coolers for different rack power configuration with and without aisle containment. The power, as well as the number of racks, are varied to study the effect of raised inlet temperature for the IT equipment in a Computational Fluid Dynamics (CFD) model developed in 6SigmaRoom with the help of built-in library items. A comparative analysis is also performed for a typical small-sized non-raised facility to investigate the efficacy and limitations of in-row coolers in thermal management of IT equipment with variation in rack heat load and containment. Several other aspects like a parametric study of variable opening areas of duct between racks and in-row coolers, the variation of operating flow rate and failure scenarios are also studied to find proper flow distribution, uniformity of outlet temperature and predict better performance, energy savings and reliability. The results are presented for general guidance for flexible and quick installation and safe operation of in-row coolers to improve thermal efficiency.

No abstract

No abstract