A Review on efficient thermal management of air- and liquid-cooled data centers: From chip to the cooling system

The steep stepwise uptake of water vapor and easy release at low relative pressures and moderate temperatures together with high working capacities make metal-organic frameworks (MOFs) attractive, promising materials for energy efficient applications in adsorption devices for humidity control (evaporation and condensation processes) and heat reallocation (heating and cooling) by utilizing water as benign sorptive and low-grade renewable or waste heat. Emerging MOF-based process applications covered are desiccation, heat pumps/chillers, water harvesting, air conditioning, and desalination. Governing parameters of the intrinsic sorption properties and stability under humid conditions and cyclic operation are identified. Transport of mass and heat in MOF structures, at least as important, is still an underexposed topic. Essential engineering elements of operation and implementation are presented. An update on stability of MOFs in water vapor and liquid systems is provided, and a suite of 18 MOFs are identified for selective use in heat pumps and chillers, while several can be used for air conditioning, water harvesting, and desalination. Most applications with MOFs are still in an exploratory state. An outlook is given for further R&D to realize these applications, providing essential kinetic parameters, performing smart engineering in the design of systems, and conceptual process designs to benchmark them against existing technologies. A concerted effort bridging chemistry, materials science, and engineering is required.

Water and Metal-Organic Frameworks: From Interaction toward Utilization.

City Energy Analyst (CEA): Integrated framework for analysis and optimization of building energy systems in neighborhoods and city districts

The continuous growth in size, complexity and energy density of data centres due to the increasing demand for storage, networking and computation has become a worldwide energetic problem. The emergent awareness of the negative impact that the uncontrolled energy consumption has on natural environment, the predicted limitation of fossil fuels production in the upcoming decades and the growing associated costs have strongly influenced the energy systems engineering work in the last decades. Therefore, the implementation of well known and advanced energy efficiency measures to reduce data centres energy demand play an important role not only to a supportable growth but also to reduce its operational costs. The carbon footprint is greatly influenced by the energy sources used. Therefore, there have been recent efforts to exploit and reuse or combine green energy sources in data centres to lower brown energy consumption and CO2 emissions. This paper presents a comprehensible overview of the current data centre infrastructure and summarizes a number of currently available energy efficiency strategies and renewable energy integration into data centres and its characterization using numerical models. Moreover it would be necessary to develop dynamic models and metrics for properly understand and quantify the energy consumption and the benefits of applying the incoming energy efficiency strategies and renewable energy sources in the data centres. Thus, the researches or investors will be able to compare with reliability the different data centre designs and choose the best option depending on the renewable energy sources and capital available.

Energy efficiency and renewable energy integration in data centres. Strategies and modelling review

Cloud computing has emerged as the leading paradigm for information technology businesses. Cloud computing provides a platform to manage and deliver computing services around the world over the Internet. Cloud services have helped businesses utilize computing services on demand with no upfront investments. The cloud computing paradigm has sustained its growth, which has led to increase in size and number of data centers. Data centers with thousands of computing devices are deployed as back end to provide cloud services. Computing devices are deployed redundantly in data centers to ensure 24/7 availability. However, many studies have pointed out that data centers consume large amount of electricity, thus calling for energy-efficiency measures. In this survey, we discuss research issues related to conflicting requirements of maximizing quality of services (QoSs) (availability, reliability, etc.) delivered by the cloud services while minimizing energy consumption of the data center resources. In this paper, we present the concept of inception of data center energy-efficiency controller that can consolidate data center resources with minimal effect on QoS requirements. We discuss software- and hardware-based techniques and architectures for data center resources such as server, memory, and network devices that can be manipulated by the data center controller to achieve energy efficiency.

Survey of Techniques and Architectures for Designing Energy-Efficient Data Centers

Aquasar: A hot water cooled data center with direct energy reuse

Hot water cooled electronics: Exergy analysis and waste heat reuse feasibility

Ever increasing device density in electronic chips is beneficial for enhancing their computing efficiency. However, it also introduces severe challenges with respect to cooling solution which are indispensible for ensuring a reliable chip operation. Such steady miniaturization will soon render the traditional air cooling strategies futile and make the switching to liquid cooling inevitable. Superior thermal properties and ubiquitous availability make water the most suitable candidate as coolant. Building up on the reported studies in the literature, which have already established the feasibility of water as coolant, here we show that the superior thermal properties of water make it possible to cool electronic chips and data centers using hot water with inlet temperature up to 60°C. The concept is demonstrated through measurements on a copper made scalable manifold microchannel heat sink and a hot water cooled data center prototype. The high exergetic efficiency achieved using hot water cooling should make it possible to reuse the heat otherwise discarded in data centers and therefore improve the overall system efficiency and lower the carbon foot print of the data centers. Finally, the encouraging results are used to model water cooling of 3D chip stacks using an interlayer integrated cooling approach. The model results are compared with measurements on a model simulator and good agreement is found, which lays the ground work for realizing a model based optimization of integrated cooling structures for 3D chip stacks.

Waste heat recovery in supercomputers and 3D integrated liquid cooled electronics

The effective refractive index of dielectric waveguides can be tuned using the thermooptic effect. In general, the tuning efficiency is polarization-dependent owing to temperature-induced stress in the layers, which causes polarization-dependent loss in optical devices. These stress issues are analyzed and tested for a high-index-contrast waveguide structure based on a silicon-oxynitride core. Experimental results are in agreement with simulations. The relative difference in tuning efficiency for transverse electric and transverse magnetic polarized light can be tuned from -3% to +3% by appropriate waveguide technology control. The optimized thermooptic phase shifters show tuning efficiency differences below 0.25%, which are reproducible from wafer to wafer.

Polarization-independent thermooptic phase shifters in silicon-oxynitride waveguides

In High Performance Computing (HPC), chiller-less cooling has replaced mechanical chiller supported cooling for a significant part of the HPC system resulting in lower cooling costs. Still, other IT components and IT systems remain that require air or cold water cooling. This work introduces CooLMUC-2, a high-temperature direct-liquid cooled (HT-DLC) HPC system which uses a heat-recovery scheme to drive an adsorption refrigeration process. Using an adsorption chiller is at least two times more efficient than a mechanical chiller for producing needed cold water. To this date this is the only installation of adsorption chillers in a data center combining a Top500 production level HPC system with adsorption refrigeration. This prototype installation is one more step towards a 100% mechanical chiller-free data center. After optimization of the operational parameters of the system, the adsorption chillers of CooLMUC-2 consume just over 6kW of electrical power to not only remove 95kW of heat from the supercomputer, but also to produce more than 50kW of cold water. This paper presents initial measurements characterizing the heat-recovery performance of CooLMUC-2 at different operating conditions.

Ingmar Meijer

Papers

Aquasar: A hot water cooled data center with direct energy reuse

Hot water cooled electronics: Exergy analysis and waste heat reuse feasibility

Waste heat recovery in supercomputers and 3D integrated liquid cooled electronics

Polarization-independent thermooptic phase shifters in silicon-oxynitride waveguides

CooLMUC-2: A supercomputing cluster with heat recovery for adsorption cooling