Researchers Calculate 10 Sextillion Processed Bytes Per Year

Jasmin Wu/UCSD Guardian

Researchers at UCSD have calculated the amount of books that data from 27 million business servers would fill: a 5.6 billion mile-high stack, repeated times 20. This is the amount of data processed by these servers per year.

According to the report, 9.57 zettabytes (or, 9,750,000,000,000,000,000,000) of information was processed in 2008, a number that researchers estimate doubles every two years. With that estimate, by 2024, the world’s servers will process information equivalent to books stacking more than 4.37 light-years from Earth to Alpha, Centauri the next star system in the Milky Way Galaxy.

International relations professor Roger Bohn said the amount of information processed is exponentially increasing. Bohn and other researchers — international relations professors James E. Short and San Diego Supercomputer Center scientist Chaitanya K. Baru — computed the workload by looking at server standards, server-industry reports, in addition to transactions, figures from information technology experts and computations that calculated the data.

The workload of the server is calculated based all information processed, though access to certain types of sites — such as YouTube — during work hours may not be necessary to business, according to Bohn.

“The type of information we are looking at now is by businesses for business purposes so in that sense, most of it is necessary” Bohn said. “It serves some business or scientific or governmental purpose- it’s work, not entertainment. Of course, some of the work that is being done is serving up web pages from things like YouTube for people’s entertainment.”

The workload of servers also varies depending on the type of server. Short said the amount of data processed by entry-level servers greatly outweighs that by high-end, more expensive servers.

“We looked at it according to workload and according to size of serves — interestingly most of the work is done by small servers, which are basically entry-level machines,” Short said. “They are doing a fraction around 2/3 of total. [High end servers] are down below 10 percent.”

According to the report, about 3.18 billion workers receive three terabytes of information per year, on average.

This figure is constantly increasing. Bohn said, with decreasing costs of computing and reduced cost of high technology, more work will occur.

“As cost of computing comes down the amount of work goes up very, very rapidly,” Bohn said. “We measured growth rates and they are on the order of 30 percent — and as high as 50 percent —compound growth rate per year.”

More to Discover
Donate to The UCSD Guardian
$210
$500
Contributed
Our Goal

Your donation will support the student journalists at University of California, San Diego. Your contribution will allow us to purchase equipment, keep printing our papers, and cover our annual website hosting costs.

Donate to The UCSD Guardian
$210
$500
Contributed
Our Goal