SBIR Phase I: Energy, space efficient, low-impedance server cooling architecture for data centers

这个小型企业创新研究（SBIR）第一阶段项目将开发一种紧凑、节能的1U服务器冷却解决方案，该解决方案使用旋转散热器或动能散热器（KHSs）来提供局部处理器冷却，同时产生冷却外围服务器组件所需的气流。该冷却技术将消除目前在1U服务器中使用的所有40毫米机箱风扇，并用产生气流的hss取代高阻抗处理器散热器。与现有技术相比，本文提出的KHS方案具有以下优点：(1)由于每个KHS产生的是压力而不是流动阻抗，因此流动阻抗更低；(2)较低的功耗，因为与40 mm轴向机箱风扇相比，hss的产流叶片更大，并且服务器流阻抗更低；(3)更小的占地面积，因为空气动器和散热器被组合成一个单一的组件。第一阶段的研究将包括组件和服务器级计算流体动力学（CFD）模拟，以及当前最先进的冷却解决方案和拟议的KHS冷却解决方案的实验验证。目标是证明与最先进的冷却解决方案相比，占地面积和功耗减少50%。该项目更广泛的影响/商业潜力源于数据中心不断上升的功率密度和总体功耗。随着“云”的爆炸式增长。数据中心的电力消耗呈指数级增长，其电力消耗是美国增长最快的电力消耗部门。组件的小型化导致单位体积产生更多的热量，而更小的产品尺寸为冷却器提供了更少的空间来散热。动态冷却是一项早期技术，它有可能显著降低冷却功率需求，同时使用更小的占地面积。目前市场上没有商用的KHS冷却解决方案；因此，拟议的KHS解决方案将是该类别中设计和销售的第一个系统。提议的服务器冷却系统的最初目标市场将是品牌和白盒服务器供应商和系统集成商。根据分析报告，2010年有760万服务器系统上线，这一数字正以4.7%的复合年增长率增长。总市场机会约为4 - 5亿美元，有可能在30个月内提供投资回报。

This Small Business Innovation Research (SBIR) Phase I project will develop a compact, energy-efficient 1U server cooling solution that uses rotating heat sinks, or kinetic heat sinks (KHSs), to provide localized processor cooling while also generating the airflow required to cool peripheral server components. The cooling technology would eliminate all 40 mm case fans currently used in 1U servers and replace the high-impedance processor heat sinks with flow-generating KHSs. The proposed KHS solution has the following advantages over existing technologies: (1) lower flow impedance, since each KHS generates pressure instead of flow impedance; (2) lower power consumption, due to the larger flow-generating blades of KHSs compared to those found in 40 mm axial case fans, combined with the lower server flow impedance; and (3) smaller footprint, since the air mover and heat sink are combined into a single component. Phase I studies will involve component- and server-level computational fluid dynamics (CFD) simulations and experimental validation of both the current state-of-the-art cooling solution and the proposed KHS cooling solution. The goal is to demonstrate 50% reduction in footprint and power consumption compared to the state-of-the-art cooling solution. The broader impact/commercial potential of this project stems from the rising power density and overall power consumption in data centers. With the explosive growth of the ?cloud? there has been exponential growth in data centers, whose electricity consumption is the fastest growing electricity-consuming sector in the U.S. The miniaturization of components has resulted in more heat being generated per unit volume, while smaller product form factors provide less space for coolers to dissipate that heat. Kinetic cooling is an early-stage technology that has the potential to significantly reduce cooling power requirements while using a smaller footprint. There are no commercial KHS cooling solutions in the market presently; consequently, the proposed KHS solution would be the first system designed and sold in this category. The initial target markets for the proposed server cooling system will be brand and whitebox server vendors and system integrators. According to analyst reports, 7.6M server systems were brought online in 2010 and this figure is growing at a CAGR of 4.7%. The total market opportunity is approximately $400-$500M with the potential for providing a return on investment under 30 months.