Statistics, multiple analysts, and research reports indicate that data centers are often overprovisioned with power and cooling capacity to maintain service levels regardless of actual IT equipment utilization. As you are well aware, this approach has proven to be expensive and inefficient. As data center energy consumption grows it is drawing the attention of CFO’s and corporate responsibility managers who are concerned with the impact of the data center’s operation on the environment and of course, the impact on the bottom line. So how can you improve your data center’s efficiency?
The path to improvement begins with an understanding of where you are and a good starting point in your effort to improve efficiency is to choose a metric that all stakeholders will accept. PUE, acronym for Power – Usage – Effectiveness and has become a widely accepted industry metric used to measure data center energy efficiency. According to the Green Grid, a PUE of 1.0 is the optimal PUE and means that 100% of the energy delivered to the data center goes to the IT equipment. (As a point of reference the EPA has the current average PUE for data centers at a 1.92, while a good target PUE to work towards is 1.2.)
So now what? You need to benchmark current or planned PUE. The best way to accomplish this is to invest in an assessment service performed by a qualified provider. Assessment typically begin with on-site data collection of power consumption data cross the data center and can also include a CFD analysis to help you visualize hot spots or underutilized cooling capacity.
Once collected, this data is used to calculate your data center’s PUE and identify particular problem areas so you can plan and implement improvements.
There are many areas within the data in which to look for energy efficiency improvements. Power and cooling capacity remain the top targets for efficiency improvement. The optimization of cooling capacity is often the most practical way for data center operators to realize short-term savings and directly impact PUE.
Speaking of practicality, some of the most fundamental steps to improve cooling efficiency is the implementation of containment systems, in cabinet ducting and floor sealing products to maximize the separation of cooled air from hot exhaust air is a proven strategy to accomplish this. The aforementioned steps are passive in nature, e.g., they do not require power consumption, there are no mechanical systems that pose potential failure points and the typical return on investment is a few months. These systems yield further effectiveness as they augment a wide range of active cooling systems that are also typically deployed in data centers. Improved space utilization is another area that can improve efficiency. While the average per cabinet heat load is 4kW, Gartner reports that 50% of data centers have cabinets with heat loads in excess of 10kW. This often results in high heat load devices being spread around the data center in underutilized cabinets to avoid creating hot spots and minimize impact on existing cooling capacity. Consolidation of high heat load devices and cabinets can help reclaim floor space and optimize existing cooling capacity.
Let’s say that you have made the investment in the “Measure” and “Improve” steps outlined above. Without a method to monitor and manage your improved energy efficiency on an ongoing basis, PUE can creep back up leading to increased operational costs as energy efficiency declines….setting you back to where you began.
Real time monitoring is the best way to provide you with visibility to conditions within the data center in order to be able to plan for, and manage changes over time. DCIM systems such as Panduit’s SmartZone™ offering collects and aggregates power and environmental data into reports enabling you to perform real time monitoring of power and environmental conditions to manage the data center power and capacity over the lifecycle of your facility.
We hope this gives you some ideas on how to approach improving your data center’s energy efficiency.