Data Center

Data centers are facilities that contain electronic equipment used for data processing (server), data storage (storage equipment), and communications networking (network equipment). Data centers also usually contain specialized power conversion and backup equipment to maintain reliable, high-quality power, as well as environmental control equipment to maintain the proper temperature and humidity for the IT equipment. They are found in nearly every sector of the economy: financial services, media, high-tech, universities, government institutions, and many others use and operate data centers to aid business processes, information management, and communications functions.

Contents 1 Operation 2 Target Areas for Energy Efficiency 2.1 Microprocessor. 2.2 Server. 2.3 Storage Devices. 2.4 Site Infrastructure Systems. 3 Reference

The demand for data centers has grown due to the following factors:

• the increased use of electronic transactions in financial services, such as on-line banking and electronic trading,
• the growing use of internet communication and entertainment,
• the shift to electronic medical records for healthcare,
• the growth in global commerce and services,
• retail moving toward real-time inventories and supply chain management
• the adoption of satellite navigation and electronic shipment tracking in transportation,
• the use of the internet to publish government information,
• the increase of government regulations requiring digital records retention,
• enhanced disaster recovery requirements,
• emergency, health and safety services,
• information security and national security,
• digital provision of government services (e.g., e-filing of taxes and USPS on-line tracking), and
• high performance scientific computing.

This demand has led to:

• growth in the shipments of servers and related IT hardware,
• rapid growth in storage technologies, including storage area networks, network-attached storage, and external hard disk drive (HDD) arrays,
• growth in the number and floor area of data centers, and
• doubling in the energy used by these servers and the power and cooling infrastructure that supports them.

It is estimated that the (data center) industry 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. This estimated level of electricity consumption is more than the electricity consumed by the nation’s color televisions and similar to the amount of electricity consumed by approximately 5.8 million average U.S. households (or about five percent of the total U.S. housing stock). Federal servers and data centers alone account for approximately 6 billion kWh (10 percent) of this electricity use, for a total electricity cost of about $450 million annually.

Forecasts indicates that U.S. consumption by servers and data centers could nearly double again by 2011 to more than 100 billion kWh, representing a $7.4 billion annual electricity cost and the output of about 25 baseload power plants. (2006)

The increase in energy use has led to:

• increased energy costs for business and government,
• increased emissions, including greenhouse gases, from electricity generation
• increased strain on the existing power grid to meet the increased electricity demand, and
• increased capital costs for expansion of data center capacity and construction of new data centers.

Thus the importance of adopting available energy efficiency solutions for data centers:

[edit] Operation

Data center rooms are filled with rows of IT equipment racks that contain servers, storage devices, and network equipment. Data centers include power delivery systems that provide backup power, regulate voltage, and make necessary alternating current/direct current (AC/DC) conversions. Before reaching the IT equipment rack, electricity is first supplied to an uninterruptible power supply (UPS) unit. The UPS acts as a battery backup to prevent the IT equipment from experiencing power disruptions, which could cause serious business disruption or data loss. In the UPS the electricity is converted from AC to DC to charge the batteries. Power from the batteries is then reconverted from DC to AC before leaving the UPS. Power leaving the UPS enters a power distribution unit (PDU), which sends power directly to the IT equipment in the racks. Electricity consumed in this power delivery chain accounts for a substantial portion of overall building load.

Electricity entering servers is converted from AC to low-voltage DC power in the server power supply unit (PSU). The low-voltage DC power is used by the server’s internal components such as the central processing unit (CPU), memory, disk drives, chipset, and fans. The DC voltage serving the CPU is adjusted by load specific voltage regulators (VRs) before reaching the CPU. Electricity is also routed to storage devices and network equipment, which facilitate the storage and transmission of data.

Cooling in data centers is often provided by computer room air conditioning (CRAC) units, where the entire air handling unit (AHU) is situated on the data center floor. The AHU contains fans, filters, and cooling coils and is responsible for conditioning and distributing air throughout the data center. In most cases, air enters the top of the CRAC unit and is conditioned as air passes across coils containing chilled water pumped from a chiller located outside of the data center room. The conditioned air is then supplied to the IT equipment (primarily servers) through a raised floor plenum. Cold air passes through perforated floor tiles, and fans within the servers then pull air through the servers. The warmed air stratifies toward the ceiling and eventually makes its way back to the CRAC unit intake.

Most air circulation in data centers is internal to the data center zone. The majority of data centers are designed so that only a small amount of outside air enters. Some data centers provide no ductwork for outside air to directly enter the data center area. Instead, outside air is only provided by infiltration from adjacent zones, such as office space. Other data centers admit a relatively small percentage of outside air to keep the zone positively pressurized.

[edit] Target Areas for Energy Efficiency

[edit] Microprocessor.

Microprocessor technology is continuously advancing, and three key trends in server microprocessor technology hold great promise for reducing server energy use in the near future: (1) the shift to multiple cores, (2) the development of dynamic frequency and voltage scaling capabilities, and (3) the development of virtualization capabilities.

Multiple-core microprocessors contain two or more processing cores on a single die, which run at slower clock speeds and lower voltages than the cores in single-core chips but handle more work in parallel (with proper software support) than a single-core chip. Additionally, because the cores share architectural components such as memory elements and memory management, signaling can be faster and consume less energy than is the case for single-core systems (Greer 2006). Published estimates on processor-level energy savings attributable to state-of-the-art multiple-core designs range from roughly 40 to 60 percent (Intel 2007, Tremblay 2006).

Dynamic frequency and voltage scaling features allow microprocessor frequency or voltage to ramp up or down to better match the computational demands. Thus, when utilization is low, the microprocessor’s clock speed can be reduced, which reduces energy consumption and heat dissipation. Frequency and voltage scaling are done automatically and constantly adjust to changes in computational demand, continuously minimizing processor energy consumption.

Virtualization allows replacement of several dedicated servers that operate at a low average processor utilization level with a single “host” server that provides the same services and operates at a higher average utilization level. Virtualization will increase the processor utilization level of the host server because of multiple virtual servers that are running and because of a small processor utilization “overhead” associated with virtualization software. Virtualization software must coordinate power-management capabilities across virtualized servers.

[edit] Server.

Major U.S. server manufacturers are clearly moving toward the production and marketing of more "energy-efficient" servers. There is the use of (1) multiple-core microprocessors with power management (i.e., dynamic frequency and voltage scaling) and virtualization capabilities, (2) high-efficiency power supplies, and (3) internal variable speed fans for on demand cooling. This is important as many volume servers, the microprocessor, cooling fan(s), and power supply losses combined can account for 50 to 80 percent of total server energy use (Dietrich 2007, Eubank et al. 2003, Patterson et al. 2006, Pouchet 2007). Advances in memory technology may also help to improve the energy efficiency as multiple-core processors and virtualization software is expected to lead to a significant increase in required server-level memory.

[edit] Storage Devices.

There is a shift to smaller form factor disk drives and increasing use of serial advanced technology attachment (SATA) drives. Additionally, improved management of storage resources may result in significant data center energy savings. Management strategies include storage virtualization, data de-duplication, storage tiering, and movement of archival data to storage devices that can be powered down when not in use. The use of solid-state flash memory devices may be an emerging energy-efficient storage option in data centers (Gonsalves 2007).

The power and cooling infrastructure that supports IT equipment in data centers uses significant energy, accounting for 50 percent of the total consumption of data centers.

[edit] Site Infrastructure Systems.

This system typically account for 50 percent or more of the total energy consumed in data centers and server installations. Improvements include low-cost measures, such as improved airflow management and optimization of temperature and humidity set points, as well as more capital-intensive measures, such as upgrading to more efficient uninterruptible power supply (UPS) systems and water-cooled chillers with variable-speed fans and pumps.

• Uninterruptible Power Supply (UPS) Systems. UPS inefficiencies can total hundreds of thousands of wasted kilowatt hours per year, but the inefficiencies can be lowered by making changes to the electrical design and/or the system configuration. Increasing UPS system efficiency offers round-the-clock energy savings, both within the UPS itself and indirectly through lower heat loads and reduced building transformer losses.

As the vast majority use a standard battery/inverter style of UPS, the simplest opportunity is a higher efficiency battery/inverter UPS unit. Some UPS manufactures are offering a high efficiency mode of double conversion UPS operation.

There are also some new technology approaches being pursued that eliminates the battery/inverter design. They may offer a longer range potential for energy efficiency. One is the rotary UPS, which utilizes a high speed, very low friction rotating flywheel usually coupled with a fast-start backup combustion generator that can start to provide emergency power. Efficiencies as high as 96% to 98% have been claimed by manufacturers of such flywheel technologies.

Another UPS alternative being explored but not yet fully embraced by the data center market is cogeneration. Use of a cogeneration plant in combination with a grid intertie can offer greater efficiency (if the waste heat is used) and inherent redundancy. Use of the waste heat is a critical issue for an efficient cogeneration system. In data centers, which typically do not have dehumidification or heating loads, the waste heat is typically used to drive an absorption or adsorption chiller to produce cooling and also to supply nearby spaces with reheat. The main hurdles to eliminating UPS systems by using a cogeneration plant plus grid intertie for redundant emergency power are first cost, the common use of natural gas to power the system, and system configurations not suited for backup operation. There is also a concern on realiability.

• Rack Level DC Power Delivery. A typical rack will contain dozens of AC to DC power supplies, each sized and provided with the equipment mounted in the rack. Utilizing a single power supply unit to perform the AC to DC conversion and supply all the equipment in a rack offers the potential to improve power supply efficiency. With this, reductions in the size of the equipment can be made. Typically, individual DC power supplies have fans to cool them; several individual fans can be replaced by a single fan. Moreover, placing a rack level DC power supply at the top of the rack can also improve the effectiveness in the removal of heat from the data center; return air grilles can be placed directly above these racks to pull the generated heat directly away and into the air conditioning system.

The multiple conversion of AC to DC within a data center powering architecture results in inefficiencies that translates into heat. The DC powering concept utilize some of the recent advancement in technologies to directly integrate DC to load such as the [www.directcoupling.com Nextek Power] router.

• Temperature Control System. Overcooling is an energy drain in many data centers. Maintains close control over inlet temperatures to eliminate wasteful cooling.

• Humiditification Control. Servers do not require tight humidity control, and can be placed in rooms with 30-70% relative humidity without adverse effects. Using this rather broad band of humidity generates immediate savings, as humidification is very energy intensive, and dehumidification can result in even higher energy costs. One of the main reasons energy use is high with decentralized humidity control is that the humidity is typically over-controlled, often with inadequate systems.

• Hot-Aisle/Cold-Aisle Configuration. Racks are arranged front-to-front so the cooling air rising into the cold aisle is pulled through the front of the racks on both sides of the aisle and exhausted at the back of the racks into the hot aisle. Only cold aisles have perforated floor tiles, and floormounted cooling is placed at the end of the hot aisles — not parallel to the row of racks.

• Centralized Air Handling System. Centralized systems use larger motors and fans, thus they have the capability for increased efficiency. These systems, when equipped with variable air volume fans, tend to have better efficiency when underloaded. However, stand-alone Computer Room Air Conditioners (CRAC) units are perceived to offer more redundancy and efficiency when compared to a centralized air handler system.

• Airside Economiser. An air handling system equipped with an airside economizer takes advantage of cool outside air temperatures, and draws this air into the data center when it is at a lower temperature than the return air coming off of the computing equipment. The heated return air is then exhausted from the building instead of re-cooled, which leads to obvious energy savings. This could cut data center cooling costs by over 60% annually.

Typical perceptions that end up serving as barriers to use of economization systems include humidity, pollutants and data center configuration. Operating in an economizer mode during period of very low outdoor air temperature (low 40°F’s or below) can result in too low a humidity in the data center, resulting to static problems. This problem can be addressed via controls, most efficiently through use of a humidity sensor to lockout economization. Pollutant concerns take many forms. Near ocean shores, concerns have been raised about salt air corroding the equipment. In some cases, when data centers are located very close to the shore, this may be a reasonable concern and should be addressed. Dust has also been raised as a concern, but standard filtration is more than adequate to address that concern.

• Cooling Plant Optimization. For large data center facilities, a chilled water system served by a central plant is the most efficient approach to providing mechanical cooling. While the chiller is not the only important element in an efficient plant design, it is the single largest source of energy consumption. Data centers offer a number of opportunities in cooling plant design and operation optimization, including: thermal energy storage, load monitoring sensors, a primary-only variable volume pumping system, and a medium-temperature chilled water loop design.

• Free cooling via Waterside Economizer. Waterside economizers use the evaporative cooling capacity of a cooling tower to indirectly produce chilled water to cool the data center during mild outdoor conditions, bypassing the energy intensive chiller. Cooling towers produces low temperature water, and a heat exchanger is used to cool the building loop while keeping it clean and isolated from the relatively dirty cooling tower water. Free cooling can provide a backup to the compressor chillers during cooler nighttime hours when plant staffing may be lower. When the weather allows, free cooling replaces the complex mechanism of a chiller with a simple, non-mechanical heat exchanger.

[edit] Reference

• Report to Congress on Server and Data Center Energy Efficiency: Public Law 109-431. US-Environmental Protection Agency (EPA): Energy Star Program, August 2, 2007.
• PG&E.com