When it comes to energy saving best practices for data centers, there are major, design-focused changes, and those that are primarily about doing things differently with what you’ve got. Implementing energy savings best practices in data centers doesn’t always have to be complicated and expensive.
Schneider Electric’s Intelligent Energy publication run an excellent overview of the “Top 12” best practices for saving energy in data centers in its March issue (pages 12-13). Some involve facility layout changes or new equipment, but other tips boil down to changes in behavior or the way that one leverages existing settings or software tools. Consider these three points from the article:
- Powering off unused equipment. It’s always best to start with the simplest fix first. Power and cooling equipment that is not needed should not be energized. Of course, data centers that have a monitoring system have a leg up when it comes to monitoring wasted energy from power and cooling equipment. APC by Schneider Electric White Paper 150 provides useful details on these monitoring and alerting functions.
- Tuning or eliminating redundant systems. Subsystems that are in place to support redundancy should be optimized for their fractional load efficiency, not for their full load efficiency. Virtualized and cloud data centers should revisit whether their redundant systems are providing any value given the dynamic nature of their loads. For those who have virtualized, make sure your power and cooling systems are an efficient match for the virtualized environment.
- Leverage the intelligence in capacity management tools. These tools help minimize “stranded capacity” in the data, allowing the maximum amount of IT equipment to be installed within the gross power and cooling envelope, which pushes the system to the highest point on the efficiency curve. For more on stranded capacity, see white paper 150, page 8.
Assuming your data center has monitoring and capacity management software tools, much can be done to pinpoint inefficiencies and improve utilization without requiring new equipment or entirely new layouts. For example, planning tools might reveal you can operate a data center at a higher power density, or with tighter safety margins. Some equipment or layout changes might be needed to achieve these goals, but the path to improvement often starts with analysis.
In other cases, heat removal equipment might have “economizer” modes that haven’t been activated because they are turned off as a default. It’s also possible that in some scenarios (but not all), there are benefits to running uninterruptable power supply equipment in “eco-mode.” For more information, see White Paper 157, “Eco-mode: Benefits and Risks of Energy saving Modes on UPS Operation.”
Analysis or professional assessment may also reveal that systems that are working at cross-purposes, such as air conditioners that are heating while others in the same room are cooling. For more details on practical strategies for reducing this type of energy waste, see White Paper 114, “Implementing Energy Efficient Data Centers,” with special attention to Table 3 on page 10.
Yes, some of the biggest energy savers are ideally “baked in” to data center design, but there is likely some low-hanging fruit. Keep an eye out for tactics such as powering off unused gear, tuning redundant systems, and taking advantage of existing eco-modes to avoid waste with the assets you have.