michelangelus - Fotolia

What does a petabyte look like?

How can you visualise a petabyte of data? What is it in real terms? Some experts try to make real the colossal size of the mighty petabyte

Just what does a petabyte look like? The US Library of Congress is often invoked when imagining a petabyte and its kin – the (smaller) terabyte and the larger zettabyte and exabyte.

Data experts at Deloitte, McKinsey, IBM, Gartner and mobile advertising firm Adfonic recently had their arms twisted to help Computer Weekly readers envisage a petabyte. 

And so, Michael Chui, principal at McKinsey, says that the US Library of Congress “had collected 235 terabytes of data by April 2011 and a petabyte is more than four times that”.

TechTarget’s own WhatIs site offers a useful point of departure for thinking about how big a petabyte is: “A petabyte is a measure of memory or storage capacity and is 2 to the power of 50 bytes or, in decimals, approximately a thousand terabytes.”

And a terabyte

“A terabyte (TB) is a measure of computer storage capacity that is 2 to the power of 40, or approximately a trillion bytes (that is, a thousand gigabytes).”

According to futurist Raymond Kurzweil, continues WhatIs, in The singularity is near, the capacity of a human being’s functional memory is estimated to be 1.25 terabytes. This means the memories of 800 human beings fit into one petabyte.

If this seems too speculative, Wes Biggs, chief technology officer at Adfonic, ventures the following more grounded measures:

  • If the average MP3 encoding for mobile is around 1MB per minute, and the average song lasts about four minutes, then a petabyte of songs would last over 2,000 years playing continuously.
  • If the average smartphone camera photo is 3MB in size and the average printed photo is 8.5 inches wide, then the assembled petabyte of photos placed side by side would be over 48,000 miles long – almost long enough to wrap around the equator twice.
  • One petabyte is enough to store the DNA of the entire population of the US – and then clone them, twice.

DVDs, battleships, and the Six Nations

Data analysts at Deloitte Analytics also put on their thinking caps to come up with the following, working up from the bits and the bytes (a bit is a binary digit, either 0 or 1; a byte is 8 binary digits long):

  • If you counted all the bits in one petabyte at one bit per second, it would take 285 million years.
  • If you counted one byte per second, it would take 35.7 million years.
  • It would take 223,000 DVDs (4.7Gb each) to hold 1Pb.
  • It would take 746 million 3.5-inch high-density floppy discs (1.44Mb each) to hold one petabyte; 746 million floppy discs weigh 13,422 tonnes (if each one weighs 18g). This is just under the size of two Type 45 destroyers, such as the newly built HMS Duncan, which has just left the Clyde.
  • Estimates of the number of cells in a human body vary, but most put the number at approaching 100 trillion, so if one bit is equivalent to a cell, then you’d get enough cells in a petabyte for 90 people – the rugby teams of the Six Nations.

Google, social media and the Big Bang

If the average MP3 encoding for mobile is around 1MB per minute, and the average song lasts about four minutes, then a petabyte of songs would last over 2,000 years playing continuously

Gartner directs attention to Google’s petabyte generation. 

“Google first entered digital mapping in 2004 and launched Google Maps and Google Earth in 2005,” says a spokesman. “Today, Google offers its users over 20 petabytes (21.5 billion megabytes) of imagery – from satellite images to aerial photos to 360-degree Street View images.”

Even back in 2008, Google was processing 20 petabytes of data a day or 7,300 a year.

Read more on the petabyte and other digital information measures

Since that time, social media sites have been generating petabytes of what has come to be called “big data”.

Facebook was storing 100 petabytes one year ago, according to its IPO filing to the US Security and Exchange Commission on 1 February, 2012.

On Twitter, there are 34,000 tweets every 60 seconds. IBM, which provides that statistic, draws attention to an astronomical project in which it has an interest: “The Square Kilometer Array (SKA), funded by 20 countries to the sum €1.5bn, is a radio telescope that can read faint signals from the Big Bang.”

The SKA (due to be completed in 2024), estimates Big Blue, will generate 1,376 petabytes per day, twice daily global internet traffic.

Finally, the Financial Times offers this animation, explaining the colossal scale of the mighty petabyte.

Read more on Data warehousing

Data Center
Data Management