Data centers house information technology (IT) equipment such as servers and network switches which are vital for our networked modern society by providing digital data storage, data processing and connectivity. Data centers house few hundreds to tens of thousands of IT equipment that consume few kilowatt-hours to multi-megawatt-hours of electrical energy that gets dissipated as heat. IT equipment need to be properly cooled so that they operate reliably for their expected lifetime. For air cooled IT equipment, manufacturers provide heat sinks, cold plates, fans, et cetera to remove heat from the vicinity of heat dissipating components and data centers need to continuously supply cold air to the IT equipment and remove hot from the vicinity of the IT equipment. Type of cooling system used in a data center is an important factor in the overall efficiency and reliability of the data center. This dissertation focuses on use of air-side economization (ASE), direct evaporative cooling (DEC), indirect evaporative cooling (IEC), and indirect/direct evaporative cooling (I/DEC) as a way to reduce cooling cost of data centers. A test bed modular data center, which has a cooling unit that operates in ASE or two-stage I/DEC modes, and located in Dallas, TX, is primarily used for this study. Included in the study are analysis of weather data to determine what percentage of a year these cooling systems can be used, modeling of the test bed modular data center using computational fluid dynamics (CFD) tool, discussion on improvements that can be made to the cooling system and factors that limit use of ASE and I/DEC, method for improving the DEC control system, et cetera. In addition, CFD modeling of another modular data center is used to show importance of proper airflow management within cold aisle of data centers and impact of hot aisle pressurization on operating point of server fans.
The skyrocketing growth in data centers, facilities that house information technology (IT) equipment for storage, management and distribution of data while striving for 24/7/365 operation with 100% up-time, has accounted for 1.3% of global energy use. A significant portion of data center energy is dedicated to removing the heat generated by IT equipment to maintain safe operating conditions and optimum computing performance. Energy efficient cooling of data center is of vital importance. One of the options for significantly cutting the cooling cost is the use of air side economization (ASE), Indirect evaporative cooling (IEC) and Direct evaporative cooling (DEC) without using expensive chilled-water systems or air-cooled CRAC units.The topology of a test bed modular data center (MDC) under consideration consists of an Information Technology (IT) module supported with a DEC and IEC module. MDC is a dynamic and complex environment with multiple mechanical and electrical control systems aimed at maintaining continuous operation of the data center. Highly nonlinear correlations, large number of constraints and multiple operating configurations make data center control a challenging research problem. In this study, a neural network model that adapts to the actual data center conditions using historical operating data and predicts the optimum configuration to reduce energy consumption is evaluated. In addition, the neural network model has the ability to learn from real time data collected from various data center sensors. When combined with a cooling unit, a predictive model that is fast and accurate in finding an optimal operating point for the modular data center unit can be implemented.In 2011, the Technical Committee (TC) 9.9 under American Society of Heating Refrigeration and Air-Conditioning Engineers (ASHRAE) expanded the operating envelope for data center thermal management in its Thermal guidelines for Data Processing environments, thus making it possible to operate IT equipment at higher server inlet temperatures and humidity and also switch to Indirect/Direct evaporative (I/DEC) and free cooling mode for increased number of hours per year. This study includes control strategies for operating I/DEC in tandem and also individually to achieve the target conditions for data center environment with minimum fluctuations in temperature and humidity. Staging of DEC for segmented cooling will provide flexibility in efficiently controlling temperature and humidity with significant water saving capability. Predictive cooling using weather forecast will counteract the start time delay of cooling modules avoiding ramping of unit due to unexpected weather conditions. The results show potential energy savings achievable through the proper implementation of control strategies and artificial neural network in operation of ASE, DEC and IEC of data center.
This book describes the use of free air cooling to improve the efficiency of, and cooling of, equipment for use in telecom infrastructures. Discussed at length is the cooling of communication installation rooms such as data centers or base stations, and this is intended as a valuable tool for the people designing and manufacturing key parts of communication networks. This book provides an introduction to current cooling methods used for energy reduction, and also compares present cooling methods in use in the field. The qualification methods and standard reliability assessments are reviewed, and their inability to assess the risks of free air cooling is discussed. The method of identifying the risks associated with free air cooling on equipment performance and reliability is introduced. A novel method of assessment for free air cooling is also proposed that utilizes prognostics and health management (PHM). This book also: Describes how the implementation of free air cooling can save energy for cooling within the telecommunications infrastructure. Analyzes the potential risks and failures of mechanisms possible in the implementation of free air cooling, which benefits manufacturers and equipment designers. Presents prognostics-based assessments to identify and mitigate the risks of telecommunications equipment under free air cooling conditions, which can provide the early warning of equipment failures at operation stage without disturbing the data centers' service. Optimum Cooling for Data Centers is an ideal book for researchers and engineers interested in designing and manufacturing equipment for use in telecom infrastructures.
Conventional data centers are extremely large buildings that have complex power distribution and cooling systems. These traditional brick and mortar data centers employ relatively expensive cooling systems and are inefficient. It has in turn led to an increase in construction and operational costs. Jonathan Koomey, Ph. D. of Lawrence Berkeley National Laboratory estimated that worldwide data center power consumption was 152.5 Billion kWh in 2005. In 2007, the US EPA estimated that servers and data centers, "consumed about 61 billion kilowatt-hours (kWh) in 2006 (1.5 percent of total U.S. electricity consumption) for a total electricity cost of about $4.5 billion. These inefficiencies of traditional data centers can be overcome by partitioning the server load into modular sections which can be deployed, powered and cooled depending on availability and requirement. Furthermore, improvements in efficiency and operational costs can be achieved by employing "free cooling" to cool the IT equipment through use of air-side economization. Air-side economizers bring in large amounts of ambient air to cool internal heat loads when weather conditions are favorable and result in substantial savings from the cost of running cooling resources. However, if ambient air properties are not suitable to cool information technology (IT) equipment directly, ambient air needs to be conditioned before entering IT equipment. One method of conditioning outside air is to use direct evaporative cooling which sprays atomized water as air passes through an evaporative cooling zone. Atomized water vaporizes and conditions air passing through the cooling zone by adding moisture and reducing its temperature, thus foregoing expensive computer air conditioning units (CRACs). The first part of the thesis will discuss various cooling techniques available for air-side economizer. The effect of various ambient environment conditions corresponding to different controls and/or environmental conditioning are required to optimize energy efficiency is studied. ASHRAE recommended Psychrometric chart for the operating environment is used to determine the water requirements for direct evaporative zone. In the second phase, computational fluid dynamics (CFD) analysis is performed using commercially available CFD tool, Fluent, to determine the performance of direct evaporative cooling in an air-side economizer. Various factors that affect performance of evaporative cooling, such as particle sizes of atomized water, ambient air temperature and humidity, water temperature are invested in this study.
Introductory technical guidance for mechanical and electrical engineers and construction managers interested in improved energy efficiency for electronic data centers. Here is what is discussed: 1. INTRODUCTION 2. INFORMATION TECHNOLOGY (IT) SYSTEMS 3. ENVIRONMENTAL CONDITIONS 4. AIR MANAGEMENT 5. COOLING SYSTEMS 6. ELECTRICAL SYSTEMS 7. OTHER OPPORTUNITIES FOR ENERGY-EFFICIENT DESIGN 8. DATA CENTER METRICS AND BENCHMARKING.
Computer cooling system design evolved over time with goals of increasing efficiency and decreasing cost. Early computers were essentially hand-built and very expensive. Reliable operation required aggressive cooling to maintain acceptable component temperatures and this was achieved with relatively low ventilation air temperatures. With time, the scale of operations increased to the point that operating cost began to strongly influence design decisions. Computer room air conditioners consumed substantial amounts of electrical power, in some situations almost as much power as the computer equipment. One cost saving idea used outside air when the ambient temperature fell below the normal cooling supply air from the computer room air conditioner. This modification acquired the term "free-cooling". Substantial cost savings from free-cooling led to the desire to expand its use to higher temperatures. Continuing to expand on this approach, some facilities ventured into evaporative cooling which proved highly successful in locations with an amenable climate. Water's latent heat of evaporation cooled the air using very little electrical power. While evaporative coolers use much less power than direct expansion units of computer room air conditioners, they have more-restrictive limitations on the allowable climate conditions of temperatures and humidity. Also, by their nature, evaporative systems use considerable quantities of water. Cooling system designers continue seeking improvements in the on-going efforts to reduce operating costs. Temperature/humidity limits for evaporative coolers are a consequence of the upper temperature limit for data center cooling supply air and thermodynamic limits of water evaporation to cool the air. Evaporative systems' cooling capacity reach a minimum during the hottest part of a 24-hour cycle. Water consumption reaches a peak at this condition as well. Designed with the necessary cooling capacity at this hot condition, the systems have excess capacity during the cooler portions of the day. Thermal energy storage offers potential to address the two negatives of evaporative cooling, restrictive limitations and high water consumption, by time-shifting cooling capacity. Thermal energy storage enables time-shifting cooling capacity from coolest portion of the 24-hour cycle when the evaporative cooler has excess capacity. Stored cooling can augment the evaporative cooler's performance at times of challenging cooling demand during the hottest portion of the 24-hour cycle. With additional cooling from thermal energy storage the data center cooling supply air temperature can be maintained in hotter environments. Cooling from a thermal energy storage system also enables the reduction of water consumption. Thermal energy storage with free cooling, when no water is used, can provide cooling later to offset water consumption. For thermal energy storage, phase change materials offer economic and performance advantages. The latent heat of phase change can store energy using much less material than sensible heat storage. The near-constant temperature energy exchange of phase change can improve the system thermal performance relative to energy storage with changing temperature. A commercially available thermal energy storage medium comes in the form of a water-based slurry of micro-encapsulated organic wax. The small, micron-size capsules in water overcome one of the major engineering challenges with many phase change materials, low heat transfer during the liquid to solid phase transition with low thermal conductivity material. While conductivity may be low, the maximum conduction distance is the capsule radius which is also small.This study investigates the benefits of thermal energy storage (TES) integrated with an indirect/direct evaporative cooler in a data center application. Concepts for integration of the TES with the cooler are developed, evaluated, and compared. Performance of the most promising candidate concept is evaluated for extended temperature operation and water conservation potential at three representative geographic locations. Capital costs for TES to be integrated with an indirect/direct evaporative cooler are estimated. Finally, operating benefits in the form of reduced operating costs are combined to determine an overall cost benefit.
Sharply increasing power dissipations in microprocessors and telecommunications systems have resulted in significant cooling challenges at the data center facility level. Energy efficient cooling of data centers has emerged as an area of increasing importance in electronics thermal management.
Recent researches and a few facility owners have focused on eliminating the chiller plant altogether by implementing 'Evaporative Cooling', as an alternative or augmentation to compressor-based air conditioning since the energy consumption is dominated by the compressor work (around 41%) in the chiller plant. Because evaporative cooling systems consume water, when evaluating the energy savings potential of these systems, it is imperative to consider not just their impacts on electricity use, but also their impacts on water consumption as well since Joe Kava, Google's head of data center operations, was quoted as saying that water is the "big elephant in the room" for data center companies. The objective of this study was to calculate the savings achieved in water consumption when these evaporative cooling systems were completely or partially marginalized when the facility is strictly working in the Economizer mode also known as 'free cooling' considering other modes of cooling required only for a part of the time when outside temperature, humidity and pollutant level were unfavorable causing improper functioning and reliability issues. The analysis was done on ASHRAE climatic zones with the help of TMY-3 weather data.
Air side economization is an arrangement of duct, damper and automatic control system which together allow introducing outside air to reduce or eliminate the mechanical cooling during mild or cold weather. In this process, ambient air is drawn inside from environment, passed through filter to get rid of contaminants and then is introduced to the cold aisle of data center. Per ASHRAE, use of air side economization is mandatory when outside air conditions are favorable to reduce data center energy consumption. To maximize economizer hours, direct evaporative cooling, indirect evaporative cooling, two stage direct/indirect evaporative cooling & compressor less cooling can be considered. The outside ambient air and heated return air upon process cooling is mixed to achieve a target cold aisle operating temperature and increases economizer hours. The dedicated space where these two air streams mix is called a mixing chamber. Mixing chamber accommodates mixing of two air streams in a short span of time. Two types of dampers are considered in this study: parallel blade dampers that rotate together in same direction and opposite blade dampers that have alternating open and shut directions. Damper blade angles have major impact on flow rate, directionality and overall mixing effectiveness. Damper blade angles for ambient air and return air inlets are varied in increments and their effect on mixed air temperature is investigated.
Free cooling ventilation is a process of using outside air to cool a building without the use of a mechanical refrigeration system. It is a very energy-efficient and environmentally friendly way to cool a building, especially in climates with cool winters and moderate summers. There are two main types of free cooling ventilation: direct and indirect. Direct free cooling ventilation: Direct free cooling ventilation systems pass outside air directly over the cooling coil of a chiller or air handling unit (AHU). This is the simplest and most efficient type of free cooling ventilation system, but it is only suitable for climates where the outside temperature is consistently cooler than the required chilled water temperature. Indirect free cooling ventilation: Indirect free cooling ventilation systems use a heat exchanger to transfer heat from the chilled water loop to the outside air. This type of free cooling ventilation system is more expensive than direct free cooling ventilation, but it can be used in a wider range of climates. Free cooling ventilation can be used in a wide range of buildings, including commercial buildings, industrial buildings, and data centers. Here are some of the benefits of using free cooling ventilation: Energy savings: Free cooling ventilation can save a significant amount of energy, especially in climates with cool winters and moderate summers. Reduced costs: Free cooling ventilation can help to reduce the costs of operating an HVAC system by eliminating the need for mechanical refrigeration. Environmental benefits: Free cooling ventilation can help to reduce the environmental impact of an HVAC system by eliminating the need for mechanical refrigeration. If you are considering using free cooling ventilation, it is important to consult with a qualified HVAC engineer to determine if it is right for your application. Here are some tips for using free cooling ventilation: Make sure the outside air is clean and free of pollutants. Use a filter to clean the outside air before it is used for free cooling. Use a heat exchanger to protect the chilled water loop from freezing. Use a control system to automatically switch between free cooling and mechanical refrigeration depending on the outside temperature. By following these tips, you can ensure that your free cooling ventilation system operates safely and efficiently.
