The way large enterprises, government agencies and managed service providers analyze and utilize energy in their data centers will undergo some fairly substantial changes over the next five years. Some organizations are already struggling with increased power demands in the face of constrained capacity caused by both inefficient equipment and stranded power. The situation is understandable, as most data centers were designed to emphasize performance and availability, with little attention paid to how efficiently the data center would deliver on those objectives.
The days of being able to take power for granted in the data center are coming to an end, however. Gone will be the costly operating model where all assets, including servers, lighting and cooling, are powered at full capacity 24×7. The new operating model will match server capacity with demand in real-time, and energy efficiency will become a key consideration for all data center assets. Here are 12 trends that data center and facility managers can expect to see beginning in 2012.
1. Organizations will have their own private clouds, and will make greater use of public cloud services for other secure cloud services as well as “spill-over” capacity during periods of peak demand.
2. Organizations will continue and even accelerate consolidation efforts within and among data centers to improve overall efficiency through greater economies of scale. An increasingly important aspect of this effort will be the energy efficiency of the individual systems.
3. Most data center operators will have a virtualized and load-balanced infrastructure. To satisfy disaster recovery needs, these virtualized and load-balanced resources will be split between at least two facilities with every application able to run in either/any facility (albeit at a potentially lower service level depending on the circumstances). An increasing number of organizations will take full advantage of this configuration by continuously load-balancing all critical applications between or among data centers, and/or grouping applications by service level priorities.
4. All aspects of monitoring and management will need to become consistent among data centers, in addition to being pervasive across the IT and facilities organizations. To achieve the best results, these ongoing efforts will require integrating the monitoring and automation capabilities, including for power, temperature and utilization. Determining the options available and having the greatest impact will require an unprecedented level of cooperation between IT and facility managers.
5. Power Usage Effectiveness (PUE) will continue to trend downwards by further reducing facility overhead, making a higher percentage of power available to the IT equipment and extending the life of data centers. As a result, the focus on IT efficiency will increase.
6. Because energy consumption tracks more closely with equipment utilization, accurate reference data for idle and peak power requirements will be needed to maximize server capacity for each data center in an energy-efficient manner. In anticipation of this need, Underwriters Laboratories created a new performance standard (UL2640) based on the PAR4 Efficiency Rating. PAR4 provides an accurate method for determining both absolute and normalized (over time) energy efficiency for both new and existing equipment.
7. To improve energy efficiency ratings, vendors are further reducing the idle power IT equipment consumes, which is making the spread between idle and loaded power consumption much wider. This will have the effect of causing more power spikes in data centers during periods of high application demand.
8. At the same time these changes are occurring in the data center, the electric grid will become increasingly unstable and, as a result, energy prices will increase and fluctuate with grid conditions. Some utilities are already beginning to charge for electricity based on its time-of-use (TOU), with rates being considerably higher during periods of peak demand, usually in the later afternoon and early evening local time.
9. Organizations will begin asking their data centers, as major users of electricity, to reduce consumption during these peak demand periods. Most data centers will do so by temporarily power-capping less critical servers and turning up the thermostat. Those with a capable Data Center Infrastructure Management (DCIM) system will be able to shift and/or shed loads within their private clouds to generate revenues from utility market participation based on the spread between actual application demand and peak application capacity.
10. Organizations will begin to participate in regional energy markets by implementing demand response, ancillary services and/or distributed generation to help stabilize the utility grid. Early innovators already have their own micro grids powered by fuel cells or other energy sources. Where renewables like solar and wind are used as distributed energy resources, reliability needs to closely managed because availability and demand can change quickly and dramatically.
11. Larger organizations will deploy multiple data centers around the globe to meet the need for growth in server capacity (in some case up to 20% annually), as well as to increase application availability and improve disaster recovery preparedness. Having multiple, strategically-deployed data centers will enable loads to be shifted to where power is currently the most stable and least expensive.
12. In an effort to gain a competitive advantage by becoming more “green” (both financially and environmentally), most companies will designate a chief sustainability officer (CSO). One of the CSO’s responsibilities will be to determine the organization’s carbon footprint, even if not required to meet any new CO2 emission regulations—another likely change that should be expected during the next five years.
Clemens Pfeiffer is the CTO of Power Assure and is a 22-year veteran of the software industry, where he has held leadership roles in process modeling and automation, software architecture and database design, and data center management and optimization technologies.