Moving common business applications – including email, customer relationship management (CRM) software, and bundled productivity software such as spreadsheets and word processing – from local IT systems to the cloud could cut energy consumption by up to 87%, a study has found.
Using cloud can save up to 23 billion kilowatt-hours of electricity, according to the six-month study by scientists at Lawrence Berkeley National Laboratory and Northwestern University, commissioned by Google.
Potential savings were seen to be most pronounced for email and productivity software, which are widely used by a large number of employees and currently rely on in-house servers that are widely dispersed and not consolidated. Using in-house datacentre infrastructure for these commonly used business applications leads to underutilisation of servers, which in turns leads to power and electricity wastage.
The project aimed to help create a model to analyse the energy and carbon impact of cloud computing services.
“We need numbers – hard data — to properly analyse how cloud computing compares with the way computing is done now,” said Northwestern University's Eric Masanet, lead author of the report. “What we found overall is that by hosting services on the cloud, as opposed to locally, the savings are pretty robust.”
Although there has been much discussion about the environmental effects of large-scale datacentres, there has been less research on the environmental benefits of providing IT services via the cloud, according to Google.
First model to assess cloud’s carbon footprint
The researchers claim that this is the first time such a model to assess cloud’s energy consumption and carbon impact is available to the public in open-access form.
The cloud energy and emissions research (CLEER) model will be useful for individuals, programme funding agencies and policy-makers as they assess computing needs and energy demands, according to Berkeley Lab.
It takes into account the factors – datacentres, transmission systems, client devices and transportation systems, among others – necessary to assess the environmental benefits or costs of shifting from local or physical resources to the cloud.
“It allows you to evaluate energy use of various options in a significant way,” said Berkeley Lab's Lavanya Ramakrishnan, the principal investigator for the project. “By studying the present-day cloud scenarios, you can see the net energy and carbon-footprint benefits at a range of scales.”
The model can be used by datacentre managers to compare current energy requirements with those that could be achieved by migrating the IT infrastructure to the cloud.
Comparisons can also be done on a regional basis, such as moving all the datacentres of a state’s universities from individual campuses to the cloud, said Ramakrishnan.
“This public use model is really a foundational tool for the energy analysis community,” said Arman Shehabi, a principal scientific engineering associate in the Environmental Energy Technologies Division.
“The analytical structure and various assumptions are fully transparent, so users can explore the model’s underlying analytics, compare different scenarios, poke at the data and discuss the results with the community,” said Shehabi.
Cloud’s energy efficiency proposition and the CLEER model to assess its carbon emissions come as Greenpeace UK urges cloud suppliers to clean up their datacentres.
Growing appetite for cloud-based IT is forcing global IT giants, including Microsoft, Apple and Amazon, to expand their cloud capacity at a breakneck pace, and this speed comes at a steep environmental expense, warned Greenpeace.
It said companies such as Amazon and Microsoft still use dirty energy resources such as coal to power their datacentres from which they provide cloud services.