InRack and InRow Cooling for Data Center

Traditional data center, computer room, and server room cooling methods cool down an entire room with cold air from centralized units to one end of the room. This approach is acceptable when power densities are minimal with few hot spots in the room. However, room-oriented designs are affected by room constraints including ceiling height, room shape, obstructions above and below the floor, rack layout, Computer room air conditioning (CRAC) units' location, power distribution, etc.

CRAC units force chilled air into a data center and around the equipment. In most cases, cooling these vast volumes of air is very inefficient. Although a hot and cold aisle containment decreases the volume, it still results in a lot of excess cooling and costs of powering the CRACs. 

So why don't consider more selective data center cooling options which work together with your CRAC units or chill specific critical loads with high efficient ways but low energy cost?

Modern data centers adopt InRack and InRow coolers (IRCs), also known as close-coupled cooling systems, because they are tailor-made for high densities of hot-running IT equipment and tight energy budgets. These cooling strategies are inherently more efficient than standard CRAC systems because it ties into the IT equipment rather than sending cooled air into the room space. They may be mounted among the IT racks / cabinets or they may be mounted overhead or under the floor.


InRack Cooling

Dedicated racks, another low-effort retrofit, offer cooling isolation. The rack operates just like a standard data center rack, but it is sealed on all sides as a self-contained system. Cool air is forced up through the rack from the bottom, running over the equipment before exiting through the top to a hot plenum, where the heat is vented or recovered as necessary.

InRow Cooling

In-row cooling systems work within a row of standard server racks. The units are standard rack height, making them easy to match with the row and couple tightly to the IT equipment to ensure efficient cooling. Systems from APC by Schneider Electric, Liebert by Emerson Network Power, Rittal and others are engineered to take up the smallest footprint and offer high-density cooling. Ducting and baffles ensure that the cooling air gets where it needs to go.

    

Compared with the room-oriented architecture, the airflow paths are shorter and more clearly defined with the close-coupled cooling systems. Smaller fans can be used due to lower volumes of chilled air; energy costs are minimized; it is easier to target air onto high-density hot spots for preferential cooling; and business continuity improves, as the failure of any one single unit in the cooling environment only affects that rack or cabinet, as opposed to the total data center or the whole aisle containment. And, as the majority of these systems are modular, it is easy and cost-effective to build in degrees of resilience, leading to higher availability across the whole data center.


About us

Strategic Media Asia (SMA) is one of the approved CPD course providers of the Chartered Institution of Building Services Engineers (CIBSE) UK. The team exists to provide an interactive environment and opportunities for members of ICT industry and facilities' engineers to exchange professional views and experience.

SMA connects IT, Facilities and Design. For the Data Center Design Consideration, please visit 

(1) Site Selection,

(2) Space Planning,

(3) Cooling,

(4) Redundancy,

(5) Fire Suppression,

(6) Meet Me Rooms,

(7) UPS Selection,

(8) Raised Floor,

(9) Code & Standards, and

(10) Transformers and Harmonic Distortion

All topics focus on key components and provide technical advice and recommendations for designing a data center and critical facilities.

Data Center Design Consideration: Cooling

Further to the previous articles in Data Center Design Consideration: Space Planning, thermal management is also an important consideration in data center design. As more cloud-based platform and applications developed and the big data trends, the demand for server usage and data storage will grow continuously. As a result, more power is consumed and more heat is generated by the servers and electronics. Proper heat dissipation techniques are therefore necessary to ensure heat is effectively removed from the equipment.

Proper design for data center cooling at least has 3 components:


Mechanical, Ventilation and Air-conditioning (MVAC) Precision System

It supplies and distributes chilled air into the computer room. The MVAC system is designed based on the total heat load of the equipment. Depending on the geographical location of your data center, temperature is typically kept between 20 - 25 degree.

Rack Layout

The TIA-942 standard recommends the use of a hot aisle and cold aisle layout in positioning racks. Racks are organised in rows with an aisle dividing each row of racks. Racks are then installed with either the front or the rear of the rack facing each other along every aisle.

The aisles, which access the front, are designated cold aisles while the aisles that access the rear are designated hot aisles.

With this layout, cooled (fresh) air is introduced into the cold aisles. As the cooled air enters the rack and moves from the front to the rear, it extracts heat from the rack-mounted equipment. At the rear, air is expelled into the hot aisles where they are directed back to the Air Handling Units (AHU) / Computer Room Air-conditioning (CRAC) Units. Using this directed path of air circulation, the computer / servers equipment are effectively cooled.



Source: http://macc.umich.edu


Rack Design

In order to make the cool aisle / hot aisle layout effective, the rack are required to be carefully installed and designed. It ensures cooled air enters the rack, is forced through the equipment and hot air is expelled from the rack:

1. Intelligent fan doors for drawing cooled air into the rack.
2. Blower fans to ensures cooled air is available along the front of the entire rack.
3. Fan doors and blower fans are used when there are insufficient air entering the cabinet.
4. Filler panels and air dams for blocking gaps between the front and rear of the rack.
5. Air diverter kits to divert cooled air upwards which helps air to flow into the equipment and to prevent hot air at the back from re-circulating to the front.

About SMA

Strategic Media Asia (SMA), a critical infrastructure training and event organizer based in Hong Kong, provides an interactive environment and opportunities for members of IDC industry and engineers to exchange professional views and experience on critical infrastructure and E&M facilities.

SMA is one of the CPD Course Providers of the Chartered Institution of Building Services Engineers (CIBSE).

For details of other data center courses and seminars, please visit our website at http://www.stmedia-asia.com/trainings.html.

Data Center Design Consideration: Space Planning

A data hall houses computers and communications equipment that delivers, store and process information. This is the heart of the data center that operates all mission-critical applications. You need to address 2 key considerations:


(1) Restricting Access

Access to the data hall shall be limited and restricted to only authorized people and to minimize any accidental damage to the computer systems.

(2) Maximizing Floor Space

Floor space in the data hall is extremely valuable due to the high investment placed into the supporting infrastructure. It includes precision air-conditioning, standby power supply, equipment-safe fire protection system as well as security and monitoring systems.



"Non-essential" Items

Sometime "non-essential" doesn't mean useless. To maximize the use and enhancing the security of the area, "non-essential" items should be located outside the data hall. Examples of these items are:


(A) Operating Console / Network Operation Center (NOC)

It houses computer stations connect to the equipment or critical facilities in a data center. NOC may be situated in a separated room / area next to the data hall. Staff can monitor and manage the various systems here and enter the data hall only when they install cables or equipment. It also ensures that the operations and maintenance of the console do not affects the racks/cabinets inside the data hall.

(B) Gas Suppression Systems

Gas Suppression Systems such as FM200 are frequently used in a data center for fire fighting purposes. Gas is used over water as they do not damage the data hall's equipment when discharging. Unlike water, gas should be stored in gas cylinders. Placing these cylinders outside the data hall would allow your technicians to serve them easier.

(C) UPS Batteries & Generators

Uninterruptible Power Supply (UPS), high voltage switchgear and Genset are critical to supply backup power and are usually isolated in different rooms / area:

• Batteries will discharge gas during charging.
• Batteries contain acid that may leak and damage equipment
• Prevention of fire
• For safety reason, only qualified electrical workers are allow to enter the room and serve the high voltage facilities

(D) Air Handling Units (AHU) / Computer Room Air Conditioners (CRACs) Units

It's commmon for a dedicated AHU / CRAC Units to be used in a data room which ensure air from other parts of the data center is not recirculated into the critical area. In addition to being bulky, the AHU or CRAC Units generates heat and maintenance issues. AHU / CRAC Units, therefore, are unsuitable to be placed inside a data hall. Should they be installed in a data hall, they should be placed along the hot aisles.

Maintenance Area

It is important equipment installed in a data center can be repaired or replaced easily without affecting normal operations. Unless we absolutely sure that rear access to the equipment is not required, no equipment should be placed along a wall such as switchgear.

A minimum space greater than the depth of the equipment is required in front or at the back of the equipment which allows

- New parts' installation
- Maintenance or repair
- Airflow (such as hot aisle or cold aisle design)

For racks, a minimum of 1m clearance is observed between racks. A 1.5m radius is usually kept at the end of cabinets / racks which facilitates the use of trolley to move heavy equipment into the aisles.



Raised Floors

Typical raised floor inside a data hall / NOC are 150mm to 300mm (or higher depends on the floor height or design request, you may refer to the international standard / advice in TIA-942 or Uptime Institute). A ramp is usually created at the entrance to bridge the different in heights. It also allows trolley to move heavy equipment. The gradient of a ramp should not be steeper than 1:12.

Per normal practices, power cables (high voltage) are usually put under the raised floor, whereas data cables are usually put inside a cable tray overhead.

Cable Routing

We suggest power cables are separated from data cables. It ensures that the power cables do not pose any electromagnetic interference (EMI) to the data cables. Per previous advice, power cables are laid under the raised floors, whereas data cables are laid overhead.

In addition, data and power cables should be entered a data hall (or other faciliities' rooms) at opposite ends of the area whenever possible to avoid too many cables on one side and EMI problem.

Layout of a data center should be designed to minimize the run of data or power cables. For example, power cables enters the data hall and are routed through

- Riser (Cable Entry)
- Electrical Distribution Box / PDU (Power Distribution Units)
- UPS (High Voltage Switches, etc.)
- Equipment Racks

with minimized distance if they are located near each other relatively.

Expansion

Data Center are used to serve enterprises, government or other organizations. As business or services developed, the demand for faster applications and data storage grow as well. You are advised to allow space for expansion purposes. This expansion space could be located near the entrance of a data center / data hall which allows easier renovation and the installation of new equipment , cabinets and racks.

Cabling, cable trays and empty racks, where feasible, can be pre-installed in the expansion areas. This would minimize the amount of work to be done and the disruption of data center operations / equipment during expansion.



About SMA

Strategic Media Asia (SMA), a critical infrastructure training and event organizer based in Hong Kong, provides an interactive environment and opportunities for members of IDC industry and engineers to exchange professional views and experience on critical infrastructure and E&M facilities.

SMA is one of the CPD Course Providers of the Chartered Institution of Building Services Engineers (CIBSE).

For details of other data center courses and seminars, please visit our website at http://www.stmedia-asia.com/trainings.html.

Invitation to Data Center Technical Visit (2 Sites) + CPD Course in Critical Facilities and Data Center Design

2-days CPD Course in Data Center Facilities Design and Infrastructure Engineering (23 - 24 October 2014)

(HKIE CPD Course Code: CPD1029)

The course is designed for Building Services Engineers, Facilities / Data Center Managers, IT Management, etc. to enrich and update the knowledge in critical facilities and data centers design & build. It is more than a general introductory program for data center. Topics include:

- IT strategy
- Cabinet layout
- Raised floor system

- Data center network and structure
- Telecommunication backbones, redundancy, sizing and planning
- Fiber and optical system design
- Fiber and optical cable components
- Copper cabling components
- Copper system design and high speed ethernet
- Cable distribution, layout and management
- Earthing / grounding and bounding

- Power (1) – high / low voltage system, switch system, etc.

- Power (2) – UPS, transformers, fuel tanks, generators, etc.

- Cooling (1) – cooling topology, hot / cold aisle, etc.

- Cooling (2) – chiller, CRAC, cooling towers, etc.

- Environmental management system
- Physical security
- Fire protection system

Date: 23 - 24 October 2014 (Thursday - Friday)

Time: 10:00 – 18:00 (14 Hours)

Venue: 20/F, New Victory House, 93 - 103 Wing Lok Street, Sheung Wan, Hong Kong

(Very Near Exit A2, Sheung Wan MTR Station)

Fee: Special rate for HKIE's all membership classes

For details and syllabus, please refer to the CPD Course Calendar of Hong Kong Institution of Engineers, HKIE (http://www.hkie.org.hk/eng/html/cpd/cpdviewer.asp?sn=3248).

Kindly complete and return an Application Form together with a crossed cheque made payable to “Strategic Media Asia Limited” - Room 1605, 16/F, Causeway Bay Plaza 1, 489 Hennessy Road, Causeway Bay, Hong Kong.

Half Day Technical Visit to Data Centers in Kwai Chung and Tseung Kwan O (8 October 2014)

Sponsored by IXTech (www.hkcix.com) and HKCOLO (www.hkcolo.net)

Further to the critical facilities course, a half-day technical visit is targeted to provide an interactive environment and opportunities for members of IDC industry to exchange professional views and experience on TWO data center facilities (Tier II and Tier III+) and services. 

Date: 8 October 2014 (Wednesday)
Duration: 13:45 - 18:00
Assembly Time: 13:45 - 14:00
Assembly Point: Exit D, Kwai Fong MTR Station (Kwai Yan Road, Next to the Bus Stop)

Fee: HK$100, Round-trip transportation will be provided
(Waiver for the 2-day CPD course's participants)

For details of the technical visit, please visit 

http://www.stmedia-asia.com/data-center-tour.html.

Invitation to Data Center Technical Visit + CPD course in Critical Facilities Design

2-day CPD course in Data Center Facilities Design
and Infrastructure Engineering (20 - 21 February 2014)

(HKIE CPD Course Code: CPD0227)

The 2-day CPD course is designed for Building Services Engineers, Facilities / Data Center Managers, IT Management, etc. to enrich and update the knowledge in critical facilities and data centers design & build.

Topics include:

- IT strategy
- Cabinet layout
- Raised floor system

- Data center network and structure
- Telecommunication backbones, redundancy, sizing and planning
- Fiber and optical system design
- Fiber and optical cable components
- Copper cabling components
- Copper system design and high speed ethernet
- Cable distribution, layout and management
- Earthing / grounding and bounding
- Power (1) – high / low voltage system, switch system, etc.

- Power (2) – UPS, transformers, fuel tanks, generators, etc.
- Cooling (1) – cooling topology, hot / cold aisle, etc.

- Cooling (2) – chiller, CRAC, cooling towers, etc.
- Environmental management system
- Physical security
- Fire protection system

Date: 20 - 21 February 2014 (Thursday - Friday)

Time: 10:00 – 18:00 (14 Hours)

Venue: 20/F, New Victory House, 93 - 103 Wing Lok Street, Sheung Wan, Hong Kong

Fee: Special rate for HKIE's all membership classes

For detail and syllabus, please refer to the CPD Course Calendar of Hong Kong Institution of Engineers, HKIE (http://www.hkie.org.hk/eng/html/cpd/cpdviewer.asp?sn=2936) or contact (852) 2117 3893.

Kindly complete and return an Application Form together with a crossed cheque made payable to “Strategic Media Asia Limited” - Room 1605, 16/F, Causeway Bay Plaza 1, 489 Hennessy Road, Causeway Bay, Hong Kong.

Half Day Technical Visit to Data Center in Tseung Kwan O
TELEHOUSE HONG KONG CCC

Sponsored by KDDI Hong Kong Limited (www.kddi.com.hk)

Further to the HKIE CPD Course, the half-day technical visit targets to provide an interactive environment and opportunities for members of IDC industry to exchange professional views and experience on critical infrastructure / E&M facilities:

Date: 12 February 2014 (Wednesday)
Time: 13:45 - 17:15
Meeting Point: Exit D, Kowloon Tong MTR Station (Please refer to a rundown attached)

Fee: HK$100, Round-trip transportation will be provided
(Waiver for the CPD course's participants)

Remark:

Business Card and HKID Card are required for the on-site registration. KDDI Hong Kong reserve the right to refuse any unexpected registration for the Technical Visit. In case of disputes, the ultimate decision will be reserved by KDDI Hong Kong.

TELEHOUSE HONG KONG CCC

TELEHOUSE HONG KONG CCC (Telehouse Hong Kong Cloud Computing Complex) is one of the largest (360,000 sq.ft) and most technologically advanced carrier neutral data center facilities in Hong Kong and in the Asia Pacific Region.

Operated by KDDI Corporation, TELEHOUSE HONG KONG CCC is a Tier 3+ purpose-built data center facility that is strategically located in Tseung Kwan O, Hong Kong. With virtually no risk of natural disasters in the vicinity, TELEHOUSE HONG KONG CCC is ideal for securing mission-critical business systems.

For detail, please visit http://www.stmedia-asia.com/data-center-tour.html

About The Blogger

Strategic Media Asia (SMA, www.stmedia-asia.com) is a leading technical training and event organizer for corporations specialized in data center design & build, E&M facilities, telecom, ICT, finance and colocation. Currently, SMA delivers a series of data center trainings and qualification programs in Hong Kong, Taiwan and Macau.

All these events / training seminars are designed to support the leadership needs of senior executives (Chief Information Officers, IT Directors / Managers, Facilities Managers, company decision makers, etc.) and to provide useful and applicable knowledge.

For detail, please visit SMA's Technical Training Seminars & Data Center Qualification Programs.

Now for Sale - Training Manual of "HVAC Design and Cooling Specialist for Data Center Efficiency"

HVAC DESIGN & COOLING SPECIALIST FOR DATA CENTER EFFICIENCY

This is a training manual developed by a team of experienced Chartered Engineers (CEng). For data center operators and facilities managers, we help you to understand key challenges and consider different factors, from design, testing, commissioning, sustainability and efficiency, of HVAC (Heating, Ventilation, and Air Conditioning) system for data centers.

In addition, the manual details the guidelines of data centers’ cooling criteria, HVAC load development, air distribution, liquid cooling, contamination, availability and redundancy, Computer Fluid Dynamic (CFD) model, energy efficiency, etc., from general theory to technical implementation.

To predict a data center’s power and heat loads over the years ahead is the hardest thing for data center owners. It introduces datacom power trends and cooling applications showing expected growth in power density for different types of computing equipment out to 2020.

Total Number of Page: 106

Total Number of Slide Contained: 204

Size: 210mm x 300mm x 10mm

Language: English

Color: Black & White

CD / Soft Copy: Not Included

THE MANUAL CONTENT

Datacom Equipment Power Trends and Cooling Applications

(a) load trends and their applications

(b) air cooling and liquid cooling of computer equipment

Design Consideration

(a) design criteria and HVAC load

(b) computer room cooling and air distribution

(c) liquid cooling

(d) availability and redundancy

(e) integration with other MEP system

(f) controls and computer fluid dynamics

Testing and Commissioning

(a) air cleanliness test, heat load test

(b) factory acceptance test and site acceptance test

(c) integrated performance test (IST)

Sustainable Design

(a) combined heat power plant (CHP)

(b) solar, geothermal and evaporative cooling method

(c) air side economizers

(d) desiccant unit

Energy Efficiency

(a) power usage effectiveness (PUE)

(b) chilled water plant optimization

(c) water side and air side equipment

(d) part load operation

(e) controls and energy management

(f) LEED certified data center

(g) building energy code

ORDER METHOD

Please contact us at info@stmedia-asia.com
The manual is also available on eBay: http://www.ebay.com/usr/st-media

How to Upgrade your Data Center and Critical Facilities?

An aging data center may no longer be able to meet the power, cooling and structural demands of advancing technologies, but few businesses have the time or the capital to build new facilities.

Fortunately, organizations can extend the working life of their data center by renovating the facility by making changes that cost little to nothing. Data center upgrades allow a business to adopt new standards and improve existing infrastructures to introduce new technologies with better performance and more efficiency.

There are several data center design changes that can extend the life of your facilities and data center


(1) Elevate your data center temperature



The data center's working temperature has long been a subject of myth and legend, but research and initiatives from industry organizations such as  the American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE) have found that data centers don't need to be cooled like meat lockers. Modern servers and other computing equipment can operate reliably at elevated temperatures.

A 2008 ASHRAE document recommended a temperature range from 65 to 80 degrees Fahrenheit for Class 1 data center equipment. Recommendations in 2011 broadened the allowable temperature range from 59 to 90 degrees Fahrenheit for enterprise-class servers and 41 to 113 degrees Fahrenheit for appropriately designed servers and other equipment.

In addition, the extended temperature range also makes it possible to adopt alternative or supplemental cooling schemes (at least during certain parts of the day), such as free air or air/water economizers -- cooling technologies that might not have even been considered when your data center was first built.


(2) Upgrade servers and systems for better consolidation and efficiency



Servers consume the majority of energy in a data center -- primarily in the processors and memory components. Organizations can gain significant energy efficiency by upgrading servers to more efficient models during normal technology refresh cycles where capital is already budgeted. The newer server may also provide greater amounts of memory, allowing a virtualized server to provide much higher levels of consolidation than earlier servers.

This means the same amount of computing work can be done with far fewer servers, saving equipment capital and generating only a fraction of the heat for a data center's cooling system to contend with.


(3) Change the system layout and rack layout for power and cooling efficiency


It is a matter of Hot Aisle & Cold Aisle.

Suppose you had a traditional data center where a large computer room air-conditioning unit (CRAC) cooled the room. Now imagine that a server refresh and consolidation project slashed the number of servers by 75%. With just a quarter of the original server count in this example, it may be possible to rearrange the remaining servers in far fewer racks and use containment to enclose the remaining servers. This limits the air volume that must be cooled, significantly reducing the amount of mechanical cooling needed and allowing for alternative cooling technologies.

In other cases, under-floor cooling may be more effective by reworking the electrical cabling, network cabling and water lines that cross below the floor.

A poorly designed and haphazard layout can obstruct cooling air distribution, making more work for the mechanical cooling unit. In addition, any water distribution increases the potential for damage to electrical and network wiring, so many organizations opt to route electrical and network wiring overhead -- leaving water lines under-floor -- and may even upgrade network cabling to allow for future bandwidth improvements.

Don't overlook the rack space itself. For example, fully populating racks can concentrate more equipment in less space, making any containment -- and associated cooling -- more effective. And some racks may not be deep enough to accommodate new generations of computing equipment. This can lead to wiring congestion and airflow problems.


(4) Consider supplemental or alternative cooling schemes



Mechanical heating, ventilation and air conditioning (HVAC) systems are a staple of the modern data center, but they are also costly, energy-hungry and a potential single-point of failure in data center availability. If the cooling system fails, a data center can overheat in a matter of minutes.

Data center renovations often focus on ways to supplement or replace traditional mechanical cooling with alternative equipment or methods that are enabled by higher operating temperatures, better containment and less equipment.

Popular alternative cooling approaches include chilled water heat exchangers (water economizers), evaporation cooling and even free air cooling (air economizers).

These methods, however, require affordable environmental resources that are suited to the task and available for much of the day. For example, using cold lake water to drive a water economizer requires a nearby lake. In many cases, these alternative methods are added to supplement traditional HVAC, lowering run times and power needs.

Organizations that must continue using HVAC are taking a fresh look at the cooling system's capacity and efficiency. The potential problem is that a large, aging HVAC system runs even less efficiently if it is used infrequently; easing the cooling load on your legacy HVAC system might actually cost more and be harder on the mechanical system.

This means that raising operating temperatures and reducing the amount of computing equipment may justify a smaller cooling system.

(5) Consider availability and reliability issues in power distribution

Upgrading the uninterrupted power supply (UPS) systems to a newer model can improve UPS energy efficiency and provide more intelligent power monitoring/measurement capabilities that complement a data center infrastructure management scheme.

When a UPS is replaced, it is hopefully with a higher efficiency system, and may also become a redundant [N+1] configuration and possibly even a modular or incremental capacity solution. Power equipment upgrades may spawn broader wiring and distribution upgrades in older buildings.

It is also a common practice to upgrade in-rack power distribution units (PDUs) to add intelligent power management, along with rack temperature and humidity monitoring. With UPS and PDU upgrades together, an organization can gather energy use data and make more informed decisions about power costs in the data center.

(6) Finally, consider the availability of data center power

Organizations with aging, unreliable or overtaxed power grids may consider local co-generation options to ensure uninterrupted power. Traditional diesel generators are quickly giving way to more efficient and environmentally friendly alternatives, including solid oxide fuel cells such as Bloom Energy Servers or solar arrays to produce some amount of local electricity. If it's not possible to install local co-generation on-site, it may be possible to contract with regional co-generation providers for supplemental electricity.

About The Blogger

Strategic Media Asia (SMA, www.stmedia-asia.com) is a leading technical training and event organizer for corporations specialized in data center design & build, E&M facilities, telecom, ICT, finance and colocation. Currently, SMA delivers a series of data center trainings and qualification programs in Hong Kong, Taiwan and Macau.

All these events / training seminars are designed to support the leadership needs of senior executives (Chief Information Officers, IT Directors / Managers, Facilities Managers, company decision makers, etc.) and to provide useful and applicable knowledge.

For detail, please visit SMA's Technical Training Seminars & Data Center Qualification Programs.

Key Areas to Examine for Data Center Efficiency

Are you struggling on where to start your energy efficiency efforts? Look to these four key areas for improvements.

• Cooling: Typically the lowest hanging fruit.
  
• Water: Don’t overlook the use of water, due to its scarcity in certain areas. Water is related to energy as well.
  
• Electrical Design: Recent engineering innovations offer new efficient options.
  
• Incentives: Help offset improvement or development costs of energy efficiency.

Measuring Efficiency

Power Usage Effectiveness (PUE) is the most popular industry metric for measuring the energy efficiency of data centers. Today, there appears to be an arms race for the lowest PUE. Even if you aren’t one of the select few with the operational flexibility to participate, you can measure your PUE and work to improve efficiency relative to your own data center site. The industry group, The Green Grid, has many resources available on PUE.

You may also consider to take some data center training / certification - Energy and Cost Management for Data Center.

Green Cooling Techniques

ASHRAE’s latest version of TC 9.9 drastically expanded the recommended and allowable temperature and humidity ranges with the approval of the major server manufacturers. It is estimated that an energy savings of 2-4% can be realized for each degree Celsius the temperature is raised in a data center. It seems raising the temperature is low hanging fruit, but I have seen very few do it to date.

Another undisputed, easy and inexpensive energy saver is hot or cold aisle containment. Preventing the mixing of cold and hot air results in a higher return air temperature that yields an increased efficiency of the cooling system. Many systems exist ranging from hard containment systems to simple refrigerator curtains that you might see in a meat locker. Have a limited budget? Hot or cold aisle containment provides a compelling financial argument for adoption.

Free cooling is now a critical consideration–with either air-side or water-side economization. The new temperature and humidity ranges offered by TC 9.9 make free cooling feasible for a large part of the year in any location, and when designing a new data center or expanding an existing facility.
Liquid cooling has been talked about a great deal of late, with liquids being far more efficient at expelling heat than air. The approach requires some modification of the server so that it can be submerged in the liquid, but studies have shown positive results.

Evaporative cooling is another energy efficient technique, especially applicable in dry climates. However, evaporative cooling often sparks a debate over the use of additional water, especially in water-constrained areas.

Water Efficiency

Water is a topic that is gaining increased attention and will continue to do so in the future. I once heard a “futurist” say that “water is the new oil.” In evaluating evaporative and other cooling techniques, many (myself included) have made the mistake of evaluating only the amount of water used in the respective cooling systems.

In order to determine the complete hydro-footprint of a system, you must also look at energy usage and how much water is used in the production of that energy. The National Renewable Energy Lab (NREL) published a study that analyzed how much water is used in the production of power per kilowatt-hour on a state-by-state basis. While not perfect, it provides a basis for analysis from an authoritative source. After taking the amount of water used in the production of energy into account in a particular geography, evaporative cooling can have a smaller hydro-footprint (use less total water) than a chilled water system due to the amount of energy saved.

There have been a few projects of late that use either sea water or ground water for cooling, which is very efficient as it effectively eliminates the need for much of the cooling equipment. A site in central Nebraska is pursuing this tactic by using irrigation wells with a volume of 1,000 GPM at 52⁰F as the source of groundwater for cooling and re-injecting the water back into the aquifer. This is not only very energy efficient, but uses little to no water for cooling, saving on both capital expenditures (CAPEX) and operating expenditures (OPEX) through the elimination of much of the cooling equipment. The net impact addresses both the energy and water efficiency of the equation for a very energy efficient, and therefore sustainable, cooling solution.

Highly Efficient Electrical Solutions

Major efficiency gains have been made in recent years in electrical equipment that can improve your data centers’ PUE. There are multiple manufacturers of UPS’ that are reaching efficiencies of 96-98 percent at less than 50 percent load. This is important if you utilize A and B feeds to your equipment for redundancy.

Another trend is for the UPS to operate in a by-pass mode, which eliminates the losses through the batteries. Many are not yet comfortable with this mode of operation, but it is another efficiency gain to consider in optimizing performance. Higher voltage and DC power are also evolving trends that provide efficiency gains that bear mentioning.

 

Energy Efficiency Incentives and Rebates

Whether designing a new, energy efficient data center or upgrading your existing facility, there are many incentives available to help defray the cost and improve your ROI.

Power companies are commonly providing incentives based upon your performance compared to a baseline building or a baseline piece of equipment. Plan to include the power company as early in the design phase as possible, to maximize the financial benefits. Some require approval of the incentive prior ordering the equipment.

Additional Considerations

There are additional considerations beyond those mentioned above in optimizing your mission critical facility’s efficiency.

• System modularity is an accepted practice that affects efficiency. Implementing modular and rapidly expandable designs in lieu of installing full density on day one typically results in higher efficiency through higher equipment utilization. This saves on CAPEX and OPEX, making for a smart business decision.
  
• Cogeneration, also known as combined heat and power (CHP), has gained in popularity and can be as high as 60-80% efficient compared to the typical 30% efficiency of normal power plants.
  
• Peak power shaving can also be achieved through thermal storage. This is done by creating ice at night when power rates are lower and utilizing the ice for cooling during the day.

Measure, Improve, Monitor and Repeat

Regardless of the selected energy efficiency measures in your new or existing data center, make sure you measure your initial or existing condition so you have a baseline. After your improvements are made, measure again to determine your new condition and your ROI. In the case of a new data center, perform a total cost of ownership (TCO) analysis to guide your decisions. You should continue to monitor your efficiency and make improvements to improve your PUE relative to your initial condition. As a reminder, measuring your outcomes against those in the industry under different operating conditions may not provide an apples-to-apples comparison.

Best Practices for Green Data Center Design & Operation

Right-sizing the physical infrastructure system to the load, using efficient physical infrastructure devices, and designing an energy-efficient system are all techniques to help reduce energy costs. A successful strategy for addressing the data center energy management challenge requires a multi-pronged approach that should be enforced throughout the lifecycle of the data center.

Best Practices - Data Center Engineering for Efficiency

System design

In datac enters, system design has a much greater effect on the electrical consumption than does the efficiency of individual devices. In fact, two datacenters comprised of the same devices may have considerably different electrical bills. For this reason, system design is even more important than the selection of power and cooling devices in determining the efficiency of a data center.

Floor layout

Floor layout has a significant effect on the efficiency of the air conditioning system. Ideal arrangements involve hot-aisle/cold-aisle configurations with suitable air conditioner locations. The primary design goal of this floor layout approach is cool air and warm air segregation.

Proper configuration of server software

When configuring servers, many data center managers are not careful about how they configure the powerrelated software. Power-economizer modes should always be selected to ensure more efficient operation of the server.

Location of vented floor tiles

In an average data center, many vented tiles are either placed in incorrect locations or an insufficient or excessive number of vented tiles is installed. By using Computational Fluid Dynamics (CFD) in the datacenter environment, the designer can optimize datacenter cool air flow by “tuning” floor tiles by varying locations and by regulating the percent of vents that are open at any given time or can optimize CRAC (Computer Room Air Conditioning) unit locations.

Rightsized physical infrastructure components

Of all of the techniques available to users, rightsizing the physical infrastructure system to the load has the most impact on physical infrastructure electrical consumption. There are fixed losses in the power and cooling systems that are present whether the IT load is present or not, and these losses are proportional to the overall power rating of the system. In installations that have light IT loads, the fixed losses of the physical infrastructure equipment commonly exceeds the IT load. Whenever the physical system is oversized, the fixed losses become a larger percentage of the total electrical bill.



Strategic Media Asia (SMA) Limited (www.stmedia-asia.com)

SMA enables individuals working in data centres to improve cost and energy efficiency. Our training seminars and workshops encompass topics associated with new data centre planning and development, and the tuning and operation of existing facilities. Service providers and equipment manufacturers will take a particular interest as the workshop takes into account the full range of technologies used in the data centre environment and its impact in how procurement decisions are made.

IT, Project and Facilities Managers, Designers, Consultants, etc. who are responsible for highly critical systems, like the UPS and CRAC systems, etc., in Data Centre and IT equipments are highly recommended to attend.

Should you have any enquiries, please visit www.stmedia-asia.com/trainings.html or www.stmedia-asia.com/events.html

Cut Data Center Energy Bill in Six Ways

"People are looking at data center efficiency, whereas five years ago it wasn't an issue," says Adam Fairbanks, Bluestone Energy, a company that retrofits old data centers to make them more energy efficient and to qualify for utility rebates (many utilities are required to help pay for data center projects that will reduce energy use; if a project can be proven to cut energy draw by 20%, the utility might pay for as much as half the cost of the project). "Today any new data center build gets scrutinized by the CFO as well as facilities and IT."


Where lowering a company's carbon footprint was a big driver for such projects a few years ago, because of the economy, environmental concerns have gotten pushed back and today they're a matter of reducing operating expense, Fairbanks says. "Money drives the majority of the projects we work on," he says.

Fairbanks shares some of the most popular methods his clients have been using to cut energy costs in a power-guzzling data center:

• Turn the thermostat up. The common wisdom around how cold a data center needs to be has changed and an ASHRAE committee has revised the upper limits of its data center temperature recommendation up to 70-77 degrees. "People have said that's conservative, and many equipment manufacturers have said that up to 90 degrees is OK for their products," Fairbanks says.
   
  However, you have to be able to manage the movement of air before you can raise temperatures, he warns. If the air is not coming through the floor properly (due to excessive wires in the way or something) or air is swirling around, you won't see efficiencies.
  
  And you still have to cool computing equipment, even with a set point of 90 degrees. A server left running by itself uncooled would probably fry itself, Fairbanks says. "At one data center I was at recently, we did a thermal scan, where we measure and map temperatures all over the facility. One rack was at 110, which is a danger level," he says. With the proliferation of blade racks, such high cabinet temperatures are becoming more common, and there's a tendency to put all the racks in one corner of the data center, which creates one huge hot spot.

• Upgrade the HVAC. "About 30% of the power used by a data center is consumed by cooling," Fairbanks says, and the average data center is over-cooled by three or four times. A new cooling system also causes less stress on day-to-day operations than bringing in other types of new equipment. "If you put in new servers and power units, you have to rewire half the data center and move things around and it's higher risk than changing the HVAC," he says. "If you have a backup HVAC system for redundancy, you can flip over to the backup while you install the new system and achieve payback quickly."

• Use cold and hot aisles. This method of laying out a data center such that cold air used to cool computers is kept separate from the hot air they generate has been around for years, but has become more widely adopted this year.

• Try blanking panels. Server racks often have holes in the back of the cabinet, especially racks that are not full of blades. The cold air that is pushed up through the floor into these rack can escape out of the holes and into the hot aisle, causing the air conditioning system to run less efficiently. A blanking panel closes over the holes so that cold air is used exclusively to cool the servers in the racks.

• Virtualize. "There's often a conflict between the business units that own the racks and the IT staff that want to use virtualization," Fairbanks says. But here's an incentive: his company has qualified data centers for utility rebates through virtualization projects, since reduced power supplies are required for fewer servers.

• Get cooling and heating equipment to work together. Some inefficiencies are caused by CRAC units that operate independently and often fight each other, Fairbanks notes. Heating systems can conflict with air conditioning and humidifiers sometimes defeat the purpose of dehumidifiers. Bluestone offers software that has sensors and controls that monitor temperature and humidity all over a data center and aggregate information from all the units to a central point that monitors and manages all the set points. The company also provides fan trays that pull air from the floor efficiently into racks where wires or other obstacles are impeding the flow of air.

Adopted from http://www.wallstreetandtech.com