Although high-energy consumption is something that we normally associate with industries such as transportation and power generation, it is surprising how much energy the global IT industry uses. In fact, a report produced for the Global eSustainability Initiative (GeSI) showed that the Information and Communications Technology (ICT) as a whole accounted for 2% of global carbon emissions in 2008, and that this will rise to 3.4% by 2020. To put this in context, this is approximately the same amount of carbon that is emitted by the global airline industry.
However, the ICT industry – which includes both IT and telecommunications – has been making significant strides to reduce its energy consumption in recent years. This doesn’t just make good environmental sense – it makes good business sense since energy is a significant component of operational costs across the industry. The same GeSI report indicated that various ICT initiatives could deliver energy savings of more than $600 billion dollars per year.
One of the biggest areas of focus for ICT energy savings is in the data center – which is understandable since the amount that the typical data center consumes is enormous. For instance, Google has estimated that its data centers consumed 2.26 billion kilowatt-hours of energy in 2010 – that is enough to keep more than 2.5 million 100W light bulbs burning for a year. There are a number of different approaches to reducing data center energy consumption, which taken together can have a radical effect.
The first approach used is known as virtualization, and is already in widespread use in large data centers. This is a software technique for getting more productivity out of each server in the data center. In the past, each major application – such as email or payroll – would have to run on a separate server. Because these applications were not used at peak capacity all of the time, the server would spend a significant amount of its time not doing anything – although it would continue to use energy. In fact, the normal utilization rate was in the 6% to 12% range. With virtualization, however, multiple applications can run on a server, so that it is used more efficiently. In some cases, this means that the number of servers can be reduced by a factor of 10.
Another important way of saving money in data centers is to reduce cooling costs. Most of the energy that goes into powering servers is translated into heat – which has to be eliminated so that equipment does not overheat. In fact, the energy needed to cool a data center can be more than the energy needed to power the equipment itself. As a result, data centers are now starting to use intelligent environmental control systems that make use of outside air when the temperature is low enough, rather than using air-conditioning. Sophisticated software is also being used to map heat and airflow in data centers, so that hot spots can be minimized – which in turn reduces the amount of air conditioning needed. Another approach involves isolating hot and cool aisles, so that air conditioning is only applied where it is needed.
Moving beyond the data center, automated workflow software is creating a greener environment in the office. These types of systems are designed to automate business processes such as customer care, manufacturing or even HR processes such as employee onboarding. One of the key environmental advantages of these systems is that they radically reduce physical paperwork and its associated environmental impact. In fact, these types of systems act as centralized repositories for digital documentation, which is delivered to users automatically when they need it. This is far more efficient from a business perspective, and it also offers the possibility of staff working from home rather than having to come into the office. This in turn reduces both real estate costs and energy consumption.
The telecommunications industry is also putting an increasing focus on energy efficiency. For example, large telcos are starting to use route optimization software to lower their energy consumption and drive efficiency at the same time. Companies such as AT&T have large field engineering forces that are responsible for installing new equipment in the field, maintaining the existing network, and repairing any problems that occur. Route optimization software monitors the location of these employees in real time, and then automatically adjusts their schedules to reduce the amount of distance that they have to travel between jobs. For instance, if a new problem arises, the closest employee who is available will be assigned, and the schedules of other employees will also be adjusted accordingly to ensure that the affected scheduled work is still completed. This approach radically reduces the amount of fuel that telcos have to use, as well as dramatically increasing their productivity.