The received wisdom has been that cloud computing is the greener option. Companies have generally reckoned that by making better use of servers, by using virtualisation and by paying for what's needed, energy consumption will be reduced.
It's true that there have been some naysayers to this - Greenpeace produced a report earlier this year decrying cloud as the least green option, but that report concentrated on how data-centres were powered rather than on the cloud data centres per se. In particular, Greenpeace deplored the practice of siting data centres in areas where they could be driven by coal-fired power stations - an objection that could be overcome by the use of more energy efficiency and carbon-neutral technology.
Australian researchers, however, have produced a study that claims that the cloud is intrinsically environmentally unfriendly. Researchers from the University of Melbourne have found that cloud is not always the cleanest option. In a departure from previous research, the academics have looked not just at the energy consumed within data centres but at the energy required to access that information -the researchers found that transporting data between data centres and users' computers, whether in the office or the home, can itself consume large amounts of energy.
In an interview with website Physorg.com, researchers' leader Rod Tucker said "Some papers that have claimed that cloud computing provides a 'greener' alternative to current desktop computing fail to include the energy consumption involved with transporting the data from the user into the cloud. In many cases, we may find that the data centre used by the cloud-based services are located in another city, state or even country.”
There's also plenty in the report about battery life of mobile consumption and the energy consumed by using the public Internet (more router hops for example). This seems a strange report. Would the power consumed by using a mobile device really exceed the power consumed by using a laptop within the home? And yes, there is a cost in receiving that data but what if you're storing your data in the energy-efficient Icelandic one mentioned above? And what about the energy savings by sending data over a mobile than in a copper cable?
No-one will deny that a lot can be done to improve data centre energy efficiency and no-one will deny that they're a big drain on energy resources - IT emits as much carbon dioxode as the airline industry according to Gartner - and that's a figure that needs to be reduced.
>But the calculations are hugely complex: the Australian researchers have taken a snapshot of the industry but it's a selective one. What about the calculation of someone accessing data from his home computer rather than travelling to his corporate offices to do so? What about the automation of processes by computer that might have needed manual intervention some years ago? The Melbourne report has been a useful reminder that other factors come into play apart from calculating data centre emissions but those variables go beyond the cost of transmitting data.
The cloud computing world is going to mean a whole ecosystem of connected servers, PCs and mobile devices. We can take a stab at cost savings and environmental impact but we're not going to get a 100 percent accurate figure - the real world is not a lab. I suspect that, in the short term, cloud computing will not be automatically the most environmentally sound approach but as servers become more efficient, as less energy is used to construct devices and as environmental factors are built into data centres, we'll start seeing a shift in attitudes.
Follow Maxwell on Twitter