TPU Inference Servers for Efficient Edge Data Centers - DOWNLOAD

Measuring power use to drive environmental sustainability at the edge

Categories Edge Computing News  |  Guest Posts
Measuring power use to drive environmental sustainability at the edge

By John Duggan, Vice President, RFCode

Sustainability is a hot topic in all industries and within corporations, the topic is focused heavily on IT. Data centers are large and complex, with environmental conditions that vary not just from area to area but also from rack to rack. Data centers consume an estimated 200 terawatt-hours (TWh) of energy each year — more than the energy consumption of some countries. Some reports estimate they also produce an estimated two percent of all global CO2 emissions.

That energy use is likely to continue to grow. Research from Transforma Insights suggests the number of edge devices deployed around the world will increase from 2.7 billion to 7.8 billion over the next decade (and that’s not including the wider category of Internet of Things (IoT) connected devices as our refrigerators and garbage bins get even smarter.)

Ensuring more sustainable practices in data centers and at the edge

With these concerns in mind, corporations need to develop sustainability plans that solve the environmental IT challenges of energy consumption. The use of edge computing reduces the amount of useless and wasteful data traveling to and from the cloud as much as possible, thus reducing overall energy requirements in the long term.

There are three core strategies to deploy that help save money while easing up the pressure put on Mother Earth.

Strategy 1: Measure power usage

You can’t change what you don’t measure. Real-time power monitoring is critical in today’s ever-changing data center environments.

Typical data center power distribution is complex. It involves many data center organizations coordinating their efforts to ensure the status quo while preserving business continuity and IT equipment uptime at the rack. Understanding the power load for each rack and its capacity is very valuable. Having these values in real-time gives data center managers the ability to fully maximize the rack’s potential.

Establishing which rack power metrics are important for a particular configuration requires information about the equipment in the rack, the site power distribution, and preference for additional functionality such as power redundancy and available power capacity. The most impactful solution is the collection and analysis of real-time, automated and granular information.

In one example, a major global bank with 35 data centers executed a consolidation initiative planning to reduce to 25 sites in order to have better power utilization and/or availability. Through the use of sensors, they were able to determine which sites had the power to spare, and which racks were under or over-utilized. The specific knowledge of power usage at the device, rack, row, and site allowed better overall space planning, saving the company money, driving site efficiency, and meeting greener initiatives.

Strategy 2: Update and locate equipment

Start with decommissioning old racks. They use too much energy and tax sustainable energy systems. Anything older than four years needs to be replaced and updated. Secondly, knowing what you have and where it is located is critical to understanding where cuts can be made. To achieve that, autonomous and real-time asset tracking, environmental, and power management sensors are key tools to know what equipment you have in racks or storage. A result of poor asset management is that servers that are not being used sometimes get missed in the decommissioning process. This results in months or sometimes even years of wasted energy.

Strategy 3: Invest in cooling technology

The right sensors allow you to safely increase set points and reduce cooling costs without risking overheating. Having a larger set of sensors for a data center room provides an earlier notification of conditions if things change. For example, 400 sensors looking at granular changes in conditions spot potential issues much faster than a room with only four sensors. The result is that data center operators do not need to arbitrarily leave a set point lower to accommodate for potential fluctuations. This also allows for a more zonal approach to isolating cooling needs.

With advanced monitoring systems in place, one leading co-location operator, Cyxtera (formerly CenturyLink), was able to safely raise set points of their edge data centers to 75 degrees, resulting in a $1.2M first-year savings and projected $15M after full global deployment. In another example, following an initiative around Energy Star compliance in their Cleveland data center facility, American Greetings mapped out a strategy to reduce energy usage and lower operating costs by turning off several air conditioning units (CRACs) that cooled the facilities. This method of running “hotter” data centers aligned well with the latest ASHRAE guidelines, but makes environmental monitoring even more important. To meet its efficiency goals, American Greetings installed advanced sensor technology that quickly provided the temperature and humidity data to help them determine the “ideal” temperature that both saved money and ensured uptime.

When environmental sensors and readers are deployed effectively, carbon footprints can be reduced. With the state of our environmental challenges worldwide today, we don’t have the time — or the power -— to waste.

About the author

John Duggan is an Information Technology and Services executive at RF Code. Duggan provides strategic leadership in data center cloud, hosting, co-location, data networking services, compliance, business continuity and disaster recovery, information security, software solutions, and related consulting services.

DISCLAIMER: Guest posts are submitted content. The views expressed in this post are that of the author, and don’t necessarily reflect the views of Edge Industry Review (EdgeIR.com).

Article Topics

 |   |   |   | 

Comments

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Featured Edge Computing Company

Edge Ecosystem Videos

Latest News