Data centers are basically the brains of any company or organization serving as a large scale storage facility. However, it looks like some of those centers are in a comatose, zombie-like state according to experts.

“A new study says that 30% of all physical servers in data centers are comatose, or are using energy but delivering no useful information,” reported Patrick Thibodeau of Computer World.

“What’s remarkable is that that percentage hasn’t changed since 2008, when a separate study showed the same thing.”

Jonathan Koomey, a researcher at Stanford University and Jon Taylor a partner at Athensis Group consulting firm, reported the statistics using data they collected from TSO logic according to the article.

Koomey commented that the zombie-like state isn’t induced by a technical error, but rather in the way these centers are managed.

“In 2008, McKinsey & Co. released its eye-opening finding that up to 30% of servers in data centers are ‘functionally dead,'” reported Thibodeau.

“The McKinsey finding was backed-up by the Uptime Institute in 2012, which also arrived at the 30% figure based on data collected from its customers.”

The servers deem themselves useless in their comatose state but proper management might be able to free the servers up and allow them to contribute to the data storage and functionality the other 70% is dishing out.

“A study last year by the Natural Resources Defense Council (NRDC), with the help of major vendors, estimated that in the U.S. alone data centers used 91 billion kilowatt-hours of electrical energy in 2013,” according to the article.

“That use is expected to increase 53% by 2020. It estimated that electrical usage could be reduced by 40% by getting rid of zombie servers and improving energy efficiency.”

Cloud centers don’t seem to be where these issues lie. According to the Natural resources Defense Council, they see the problem being more prevalent in smaller data centers. Both Taylor and Koomey expect to continue their work to find more evidence of these zombie servers by working closely with TSO logic.

Read the full story.

Share and Enjoy !

0Shares
0 0