Ashrae Journal - November 2008 - 25

Opportunities for Savings The sequential approach outlined in this article prioritizes the most effective energy reduction strategies, but it is not intended to be a comprehensive list of energy reducing measures. In addition to the actions in this approach, data center managers should consider the feasibility of the following: • Locate and disconnect ‘ghost’ servers. These are servers that are no longer performing any useful tasks as their loads have been moved to newer servers, but are still connected and consuming energy. • Consolidate data storage from direct-attached storage to network attached storage. Also, faster disks consume more power so consider reorganizing data so that less frequently used data is on slower archival drives. • Use economizers where appropriate to allow outside air to be used to support data center cooling during colder months, creating opportunities for energyfree cooling. With today’s high-density computing environment, economizers can be cost effective in many more locations than might be expected. • Monitor and reduce parasitic losses from generators, exterior lighting, and perimeter access control. For a 1 MW load, generator losses of 20 kW to 50 kW have been measured. high efficiency power supplies. Facilities that experience predictable peaks in activity may achieve the greatest benefit from power management technology. Figure 5 shows how compute load and type of operation influence priorities. Actions Needed to Further Efficiency The research on this sequential approach has highlighted the need for additional work necessary to further the industry’s pursuit of maximizing data center efficiency. What follows are the most significant areas requiring attention. Deἀne universally accepted metrics for processor, server and data center efἀciency. There needs to be an easily understood and easy-to-use measure, such as the miles-per-gallon automotive fuel efficiency ratings, which can help buyers select the ideal processor for a given load. The performance per watt metric is evolving gradually with SPEC score being used as the server performance measure, but more work is needed. This same philosophy could be applied to the facility level. An industry standard of data center efficiency that measures performance per watt of energy used would be extremely beneficial in measuring the progress of data center optimization efforts. The PUE ratio developed by the Green Grid provides a measure of infrastructure efficiency, but not toNovember 2008 tal facility efficiency. IT management needs to work with IT equipment and infrastructure manufacturers to develop the miles-per-gallon equivalent for both systems and facilities. More sophisticated power management. While enabling power management features provides tremendous savings, IT management often avoids this technology because the impact on availability is not clearly established. As more tools are created to manage power management features, and data is available to ensure that availability is not impacted, we should see this technology gain market acceptance. More sophisticated controls that would allow these features to be enabled only during periods of low utilization, or turned off when critical applications are being processed, would eliminate much of the resistance to using power management. Matching power supply capacity to server conἀguration. Server manufacturers tend to oversize power supplies to accommodate maximum configuration of a particular server. Some users may be willing to pay an efficiency penalty for the flexibility to more easily upgrade, but many would prefer a choice between a power supply sized for a standard configuration and one sized for maximum configuration. Server manufacturers should consider making these options c c c c c c c c c c c www.info.hotims.com/16019-18 ASHRAE Journal 25

Ashrae Journal - November 2008

Table of Contents for the Digital Edition of Ashrae Journal - November 2008