Companies that design new datacentres with a focus on energy efficiency and effective utilisation of space are seeing growth in capacity of as much as 300 per cent with 60 per cent less space, according to industry analyst Gartner.
This is largely due to the fact that most IT assets are currently being underutilised in old datacentres, according to the analyst firm, which points out that some servers are running at as little as 12 per cent productivity.
It also said that average racks of old datacentres are populated to just 50 or 60 per cent of capacity, with floor space spread out for cooling purposes.
"There is a real and growing desire to increase productivity in datacentres," said Dave Cappuccio, chief of infrastructure research at Gartner.
"Organisations are starting to take a serious look at consumption ratios of compute power to energy consumed and then compare them against estimated productivity of applications and the equipment to deliver the applications," he added.
Companies are now being forced to do a u-turn on spreading physical infrastructure for cooling purposes, as more servers are required, racks are becoming densely populated and floor space is at a premium. This inevitably pushes up power and cooling costs.
Gartner suggested that this trend is likely to result in operating costs of servers being so high that they could easily match capital costs within the first few years of investment.
"The days of idle machines sitting on the datacentre floor during off-peak hours will be a thing of the past. At current energy rates a 40kW rack could cost upward of £3,300 per server, per year," Cappuccio explained.
"Organisations need to make a break with the past and realise that innovation in datacentre design will yield both reduced capital and operating expenditure.
"Think small, think dense – the objective is the highest compute performance per kilowatt," he added.
Gartner outlined a number of methods companies can implement to counter power consumption and improve efficiencies.
Row- and rack-based cooling can reduce energy consumption by 15 per cent, while rightsizing a new datacentre and only expanding when needed can reduce long-term operating expenses by 10 to 30 per cent.
Gartner also said that companies should be virtualising as much as possible.
By eliminating high entry costs for big data analysis, you can convert more raw data into valuable business insight.
A discussion of the "risk perception gap", its implications and how it can be closed