How Green is your Data Center?

Kobi Haggay is Vice President of Products and Marketing for RiT Technologies.

The demand for data center operations and energy costs are rising. Global data center traffic is estimated to grow threefold from 2012 to 2017 and power requirements will increase as well according to Devva Bodas, lead architect for The Green Grid Association.

Many organizations are announcing green initiatives. Apple announced that its massive data centers in Ireland and Denmark will be powered entirely by renewable energy, and Amazon is investing in an Indiana wind farm. But it's not just the huge enterprises that feel the pressure to be energy efficient. Specifications for communications rooms and data centers are also challenging because of tougher space, power and environmental constraints; and reducing energy costs requires more precision than ever for companies of all sizes.

Here are some measures you can take to reduce your carbon footprint.

You Can’t Manage what You Can't Measure

The first step in greening your data center is understanding, in detail, what you have, what you use and don't use, and why. In order to do so, companies need to take inventory of their data center resources. Everyone wants to skip over this task because it’s tedious. There is the temptation not to “waste” time and money gathering data and analyzing it and to do something immediately to make things better. But you can’t make informed decisions unless you can show on paper what you have, what you use, and how much it costs.

There are tools that can provide complete real-time control of all network physical components and their connections. The AIM (automatic infrastructure management) standard, due to be released later this year, describes systems that will enable the automatic documentation of all network components and their connections resulting in a better-managed network with greater energy efficiency.

The idea behind AIM is simple yet powerful: to enable real-time management of the network infrastructure by using self-aware network components, a central data repository and intelligent processes. To accomplish this task, IP discovery is used to build an accurate topological map in real time.

Everything can be documented in the database including racks, servers, switches, routers, patch panels, storage, PDUs, power strips, UPSs, switchgears, and switchboards, etc. This information can be stored at a high level of granularity where all ports and all slots are described and named to optimize space in equipment rooms and space in racks. Configurable devices, such as enclosures, can be managed with a high degree of accuracy showing the exact configuration, and making sure that the right blades are put in the right places.

These systems can also be used to identify underutilized components for better capacity management. When capacity isn't optimized, it results not only in higher energy costs, but it can also contribute to longer response times and lower productivity.

In addition, more accurate and complete documentation reduces the labor required to repair, move or change devices. Human error, such as miscalculations, incorrect port locations or mispatches, lead to wasted resources and unnecessary downtime.

Even more importantly, AIM can support remote network moves, adds, and changes that eliminate travel time and expenses.

Monitoring for Energy Efficiency

The same system can be used to monitor temperatures, including return water for cooling systems to reduce energy costs. Real-time monitoring of power and energy ensures that cooling matches the real need.

In addition, by monitoring when IP addresses are used, it is possible to pinpoint underutilized work areas and to dim lights and reduce cooling or heating requirements accordingly. This increased level of surgical energy deployments can result in significant savings.

According to Gartner, intelligent infrastructure management in the data center can cut operational costs by 20 to 30 percent, including optimized power and space utilization. This is why AIM platforms are advocated by industry leaders as a new best-practice for managing data centers.

It may be difficult to compete with Amazon and Apple when it comes to energy efficiency. However, when the planning and organization of the data center is supported by AIM, data center managers can start with a complete and accurate knowledge of what they have. This level of precision can only result in reduced energy requirements and more efficient operations.

Industry Perspectives is a content channel at Data Center Knowledge highlighting thought leadership in the data center arena. See our guidelines and submission process for information on participating. View previously published Industry Perspectives in our Knowledge Library.