12 Data Center Energy Management Trends for 2012 and Beyond
12 Data Center Energy Management Trends for 2012 and Beyond
The days of IT managers being able to take power for granted in the data center are rapidly coming to an end. Beginning in 2012, the way organizations -- enterprises, government agencies and service providers alike -- all analyze and utilize energy in their data centers will witness some major changes.
Already, a growing number of organizations are starting to grapple with increased power demands in the face of constrained capacity. The reason underlying the need to make these changes is rather simple: Most data centers were designed for high performance and high availability, with little or no attention paid to power.
Here are 12 changes data center and facility managers can expect to experience over the next few years beginning in 2012.
1. Most organizations will have a private cloud, and will make greater use of public clouds for “spill-over” capacity during periods of peak demand. This change is already underway for what is becoming something data center managers will be pursuing far more frequently: efficiency -- in this case from the economies of scale afforded by cloud-based infrastructures.
2. To further improve efficiency, organizations will continue and even accelerate their server and storage consolidation initiatives within and among data centers. Increasingly important in these efforts will be the elimination of stranded power and the energy efficiency of individual systems.
3. The consolidated infrastructure will be more virtualized and the load more fully balanced. To improve disaster recovery, the virtualized and load-balanced resources will be shared between at least two facilities with every application able to run in either/any. An increasing number of organizations will take this configuration to the next level by continuously load-balancing all critical applications between or among data centers.
4. All aspects of monitoring and managing the data center infrastructure will need to become consistent, in addition to becoming pervasive across the IT and facilities organizations. Achieving the best results will require a holistic view of power consumption, environmental conditions and resource utilization. Making the optimal tradeoffs to have the greatest impact will require an unprecedented level of cooperation between IT and facility managers.
5. Power Usage Effectiveness (PUE) will continue to improve by further reducing facility overhead (especially for cooling), which will have the dual effect of making a higher percentage of power available to the IT equipment and extending the life of the data center. A related change will be a migration from the costly operating model of powering all assets (including servers, lighting and cooling) at full capacity 24x7, to a new model that matches server capacity with demand in real-time. The improvement in PUE also means there will be far more focus on the IT efficiency portion of the equation.
6. Because energy consumption tracks closely with equipment utilization, IT managers will need accurate reference data for idle and peak power utilization to maximize server capacity in an energy-efficient manner. In anticipation of this need, Underwriters Laboratories created a new performance standard (UL2640) based on the PAR4 Efficiency Rating, which provides a very accurate method for determining both absolute and normalized energy efficiency for both new and existing equipment.
7. Vendors have also noticed this increased focus on energy efficiency, and are making changes, including reducing the idle power consumed by IT equipment. Unfortunately, this means that the spread between idle and loaded power consumption is increasing, and that will cause more power spikes to occur in data centers during periods of high demand.
8. Concurrent with these changes in the data center are some in the electric grid. The grid is becoming increasingly unstable and, as a result, energy prices are beginning to increase and fluctuate with grid conditions. A few utilities are already setting rates based on time-of-use (TOU), with those during periods of peak demand (usually in the later afternoon and early evening) potentially being considerably higher.
9. Data centers will become a primary candidate for reductions during peak demand periods. Data centers consume a significant amount of electricity in any organization, so companies' basic response will be to temporarily power-cap less critical servers and turn up the thermostat. More granular responses, such as shifting and/or shedding loads within the private cloud, are possible only with a capable Data Center Infrastructure Management (DCIM) system. A few organizations may even opt to participate in utility market programs to generate revenue from the spread between actual application demand and peak application capacity.
10. Utilities are beginning to roll out other programs that enable organizations to participate in regional energy markets by implementing ancillary services and/or distributed generation -- both of which also help stabilize the grid. Some early innovators already have their own micro grids powered by fuel cells or other energy sources. But be careful with intermittent power from solar or wind, as the difference between availability and demand can change quickly and dramatically.
11. Larger organizations will take full advantage of having multiple, strategically-located data centers to shift loads to where power is currently the least expensive and/or more stable. This approach has some additional advantages, as well, including increasing application availability, improving disaster recovery preparedness, and satisfying the growing demand for server capacity (in some case up to 20 percent annually).
12. Most businesses will designate a chief sustainability officer (CSO) as a means to become more “green” both environmentally and financially (There are, after all, genuine savings in being efficient.). Among the CSO’s responsibilities will be determining the organization’s carbon footprint, whether or not this is required by some new CO2 emission regulation. And that’s another change that might be expected as early as 2012.