Experts Estimate Spy Data Centre Capacity

After it was revealed that American spooks are spying on the online activities of people around the world, a new government database in Utah is now in the centre of attention.

The database is supposed to be a place to house and process information gathered from phone and Internet service provider, satellites and fiber-optic cables. The centre is expected to open in September and hold “yottabytes” of information or five “zettabytes” according to different sources. Anyway, it is more data than human brains are able to comprehend.

However, Forbes has started to work out that the actual storage capacity of the new database is comparatively low. After factoring out all the administration buildings, power sources and back-up generators, there is room for 100,000 square feet of servers – though impressive, still not enough for a yottabyte of data.

According to Brewster Kahle, the engineering genius behind the Internet Archive, a space of that size is estimated to hold 10,000 racks of about a $1 billion worth of servers. If each rack can store 1.2 petabytes of data, and annual voice recordings of all the US phone calls take up 272 petabytes, it will make it just over 200 of those 10,000 racks. As such, database can hold up to 12 exabytes.

In the meantime, Internet infrastructure expert Paul Vixie claims that those calculations are a little on the high side, because if larger 13 square feet racks are used, with a lower amount of data storage per rack, it will make it less than 3 exabytes of data capacity for the database. This will allow to store 24-hour recordings of all Philadelphia’s residents per year.

The US authorities justify their effort by saying that they are only collecting metadata. However, the experts point out that this data is able to tell a lot about people and all of their social connections and habits, including who they talk to and where they go. Of course, it is unlikely that the government stores every communication, but it still can help the state decide who, when and where to target before zooming in on them via other back doors.

As the numbers are speculation, to put it into perspective, the new database would still hold plenty of information. For example, Google uses multiple data centers for a single exabyte of data.