6.1.1    All rack layouts should incorporate hot-aisle/cold-aisle configurations with strategically placed air conditioner vents located in the raised floor tiles. The primary design goal of this layout is the segregation of cool air and warm air to maximize cooling efficiency. The rack layouts fall under the responsibility of the Computer Operations Facilities staff.

6.1.2    The Computer Operations Facilities staff is responsible for proper location of the vented floor tiles to insure the room is balanced correctly for maximum cooling capacity. Worldwide Environmental Services (WES) or another contracted party will conduct three site visits per year to optimize the Data Center cool air flow by “tuning” floor tiles. This includes varying locations and regulating the percent of vents that are open at any given time.

6.1.3    Managers responsible for projects that include new servers, storage and networking equipment should perform research to predict future power and cooling loads. Preference should be given to equipment earning the EPA’s ENERGY STAR designation. The use of variable speed fans for on-demand cooling should be utilized when possible. Equipments’ power supplies vary greatly in the amount of energy consumed; utilizing better power supplies which more efficiently convert AC power at the plug to DC power used by the internal components can have a significant impact on the energy consumed over the lifetime of the equipment. The American Society of Heating, Refrigeration & Air Conditioning Engineers (ASHRAE) offers a series of design guides that may assist with your research, along with the assistance of the DISC Data Center Manager.

6.1.4    System Administrators should be thoughtful in how they configure the power-related software on servers.  Power-economizer modes should always be selected to ensure more efficient operation of the server.

6.1.5    Installation of blanking panels on racks to improve airflow dynamics is recommended. These panels cover sections of racks that do not have equipment installed. The panels are an inexpensive way to decrease server inlet temperature while increasing the Computer Room Air Conditioner (CRAC) return air temperature, thus maximizing the efficiency of those units.

6.1.6    The selection of new hardware should take into consideration the “carbon footprint” of the equipment. This includes the energy consumed during the production, shipment, and operation of the equipment over its lifetime. Ask potential vendors to provide a value representing their carbon footprint of the manufacturing process and projected energy consumption when in operation. According to a Gartner, Inc. study, the carbon emitted before you even hook up your equipment can be as much as 80 percent of the energy used by the system. Consider giving preference to equipment manufactured by vendors who are members of The Green Grid organization (WWW.THEGREENGRID.ORG), a consortium of information technology companies and professionals seeking to lower the overall consumption of energy in data centers around the globe.

6.1.7    Equipment used for Disaster Recovery can be a significant source of energy consumption. To get the most use of available power, consider running your Development and Test applications on your Disaster Recovery infrastructure to maximize efficiency.

6.1.8    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. Storage managers should consider why infrequently-used data needs to reside on high-speed disk devices; rarely used data could be stored on slower drives or even on tape, both of which would consume less power and energy.

6.1.9    Virtualization consolidates existing and expected future workloads to reduce the number of physical servers required, thereby reducing floor space, power and cooling requirements. It may also increase the utilization of servers to improve energy efficiency and place more computer capability into production. The widespread underutilization of servers is one of the most often-cited reasons for suboptimal energy efficiency in data centers. Maximizing the utilization of existing servers therefore represents one of the most significant opportunities for energy savings.

6.1.10    “Right-sizing” can also contribute to the high efficiency of the Data Center. Below are examples of how to leverage efficiency during server consolidation:

• Use a two-way server or a single-processor dual core server to replace two or more old servers.
• Replace an old server with a blade based on a low-voltage or mid-voltage processor.
• Replace a dual-processor server with a single, dual-core server with a single, dual-core processor.
• Use a two-way dual-core server in place of a four-way server.
• High-speed memory contributes to server energy consumption. Architect your servers with the proper amount of memory to accommodate the intended application.  Use application modeling tools or vendor recommendations when in doubt.

6.1.11    Eliminate servers and other equipment that are no longer in use. Turning off “comatose” servers and storage can yield immediate savings. Those are defined as components which are installed and running, but which are running applications no longer needed or which run no applications at all. It is not uncommon to leave equipment installed and powered up after replacement or retirement. This is an often overlooked opportunity for savings.

6.1.12    Some computing equipment contains substances which can be hazardous to the environment if not disposed of properly. Some vendors are addressing this problem by reducing the amount of hazardous materials contained in their products. For example, Sun Microsystems advertises that some if its servers are RoHS compliant, meaning they comply with the “Restriction of Hazardous Substances” such as lead, cadmium, mercury, hexavalent chromium, polybrominated biphenyl (PBB) and polybrominated diphenyl ether (PBDE) flame retardants. Many vendors have similar efforts underway. Consider giving preference to those vendors taking this initiative.

6.1.13    Proper disposal of computing equipment has an impact on the environment due to the inclusion of hazardous materials mentioned above. At this time few organizations address this issue, and fewer disposal/recycling companies will guarantee proper handling of such materials. Some groups have proposed legislation requiring specific disposal methods for computing equipment, and some have even proposed tracing the serial numbers of improperly disposed equipment to the owner for prosecution. Maintaining credible documentation is key to proving that proper procedures were followed in the disposal of your equipment. The State Surplus Property division is a commonly used disposal service. Eligible computers and peripherals with a useful life are offered for sale to other agencies and the public. Computers and peripherals without a useful life are then sent to Asset LifeCycle, LLC. Asset LifeCycle is a permitted E-Waste Recycling and Reclamation Facility in the State of Kansas, and can handle disposal of any equipment that Surplus Property can not accept. Of course, you are responsible for cleansing your equipment of all data before disposal, which is in itself a serious matter.

Summary:

Clearly, any energy efficiency initiative should not be allowed to reduce IT availability or the quality of service that the data center provides. However, decision makers will likely find often-overlooked efficiency opportunities if they remember to consider what new configurations will better serve the enterprise business needs, rather than just repeating previous practice in a slightly better way. While the most immediately obvious “greening” techniques are those related to energy savings once your equipment is up and running, don’t overlook the environmental impact of manufacturing, shipping, and disposal of equipment.

General Items:

6.3.1    The DISC Computer Operations Facilities staff acts as a liaison between IT and the Division of Facilities Management. This organizational alignment facilitates an energy-efficient Data Center policy that addresses both IT systems and physical infrastructure systems.

6.3.2    Computer Operations Facilities staff are responsible for Data Center issues specific to hardware planning, electrical deployment, heat removal and physical Data Center monitoring. The staff should always be consulted on these issues pertaining to the Data Center.