While cloud providers have long touted the cost savings of using cloud services to customers, new research shows that US businesses who switch to the cloud actually help save a significant amount of energy as well.
The potential for energy savings is substantial, according to the case study by researchers at the Lawrence Berkeley National Laboratory. The results show that if all US business users shifted their email, productivity software and CRM software to the cloud, the primary energy footprint of these applications might be reduced by as much as 87 percent or 326 Petajoules.
According to the Google-sponsored case study, of the 86.7 million US workers who use computers, approximately 58.9 million regularly use productivity software applications like word processing, spreadsheets and file sharing. Potential savings are especially pronounced for email and productivity software, which are used by a large number of employees and currently rely on servers that are widely dispersed in mostly smaller data centers where servers are underutilized. Obviously, these results bode well with Google as it tries to sell Google cloud services onto business users. The results do provide an interesting talking point for cloud providers talking to customers about the possible benefits of cloud.
“If all present-day systems would fully shift to the cloud, our results suggest that the primary energy footprint of US business email, productivity and CRM software could be reduced to around 47 PJ each year,” according to the study. “The operational energy use of data centers is significantly reduced when moving from local data centers to the cloud. This result is not surprising, given that the cloud is expected to use far fewer servers in far more efficient data centers compared to present-day local data centers.”
“Our results also indicate smaller, but nontrivial, reductions in the embodied energy of data center IT devices in the cloud given that far fewer servers are required for the same computations. Lastly, we predict a small increase in the operational energy use of data transmission systems when shifting to the cloud, which can be attributed to increased data traffic associated with remote software access,” researchers say.
The researchers created a model, using Google AppEngine, to analyze the potential for energy savings associated with migrating US business software to the cloud, called the CLEER Model. The model estimates the net energy use and emissions of data center services across “all major societal end uses of energy.”
Researchers hope that the transparency of the CLEER Model combined with the case study will promote other researchers and decision makers to build “more comprehensive and impactful analyses” on the net energy and emissions implications of cloud services.
Last year, the New York Times ran a feature series on data center efficiency, which many in the data center industry felt was misguided and outdated, and was missing the narrative of how the industry has responded to the challenge of inefficiency.
What do you think of the results of the Google case study? Do you think that cloud actually can help save energy? Let us know in a comment.