Note: This post was originally published at Oulixeus Consulting
A lot of news articles recently (Google News currently shows 1,060 articles) are citing the annual EMC-IDC Digital Universe studies of the massive growth of the digital universe through 2020. If you have not read the study, it indicates that the digital universe is now doubling every two years and will grow
44-fold 50-fold now 55-fold from 0.8 Zettabytes (ZB) of data in 2009 to 35 40 now 44 Zettabytes in 2020. (Every year IDC has revised the growth curve upward by several Zettabytes.)
Usually these articles show a diagram such as this:
This type of diagram is great at showing how much 44-fold growth is. However it really does not convey how big a Zettabyte really is—and how much data we will be swimming (or drowning in) by 2020.
A Zettabyte (ZB) is really, really big – in terms of today’s information systems. It is not a capacity that people encounter every day. It’s not even in Microsoft Office’s spell-checker, Word “recommended” that I meant to type “Petabyte” instead
The Raw Definition: How big is a Zettabyte?
A Computer Scientist will tell you that 1 Zettabyte is 270 bytes. That does not sound very big to a person who does not usually visualize think in exponential or scientific notation—especially given that a one-Terabyte (1 TB) solid state drive has a capacity to store 240 bytes.
Wikipedia describes a ZB (in decimal math) as one-sextillion bytes. While this sounds large, it is a hard to visualize. It is easier to visualize 1 ZB (and 44 ZBs) in relation to things we use everyday.
Visualizing Zettabytes in Units of Smartphones
The most popular new smartphones today have 32 Gigabytes (GB) or 32 x 230 bytes of capacity. To get 1 ZB you would have to fill 34,359,738,368 (34.4 billion) smartphones to capacity. If you put 34.4 billion Samsung S5’s end-to-end (length-wise) you would circle the Earth 121.8 times:
Click to see a higher resolution image and the dot that represents Earth to-scale vs. the line
You can actually circumnavigate Jupiter almost 11 times—but that is not obvious to visualize.
The number of bytes in 44 Zettabytes is a number too large for Microsoft Excel to compute correctly. (The number you will get is so large that Excel will cut off seven digits of accuracy–read that as a potential rounding error up to one million bytes). Assuming that Moore’s Law will allow us to double the capacity of smartphones three times between now and 2020, it would take 188,978,561,024 (188+ trillion) smartphones to store 44 ZB of data. Placing these end-to-end- would circumnavigate the world over nearly 670 times.
This is too hard to visualize, so lets look at it another way. You could tile the entire City of New York two times over (and the Bronx and Manhattan three times over) with smartphones filled to capacity with data to store 44 ZBs. That’s a big Data Center!
Amount of Smartphones (with 2020 tech) you would need to store 44 ZB (click for higher resolution)
This number also represents 25 smartphones per person for the entire population of the planet. Imagine the challenge of managing data spread out across that many smartphones.
Visualizing Zettabytes in Units of Facebook
Facebook currently stores 300 Petabytes (PB) or 300 x 250 bytes of data in their data warehouse (this is growing at 600 TB per day). This is rather enormous—especially for a private enterprise. However it is much, much smaller than even one Zettabyte: 1 ZB could contain 3,495 entire Facebook Data Warehouses:
1 ZB would contain 3,495 entire Facebook Data Warehouses (click for zoomable resolution)
Facebook leadership has expressed the desire to get everyone on Earth connected. Today, Facebook has 1.3 Billion Monthly Active Users. If they got all 7.5 billion people projected to be alive in 2020—and data continued to grow 10-fold between now and then—their data warehouse would still only be 1/60th of 1 ZB:
7.5 billion MAUs AND a 10x increase in data use is still only makes the DW 1/60th of a ZB
Expanding this to 44 ZB increases the number to 153,791 Facebook Data Warehouses. A grid display of this number is beyond the resolution of even Retina Display monitors. You could visualize this with “Bat Cave-like” grid of four rows of 11 hi-resolution monitors. However, that is impractical and expensive (~$40,000).
Really Big No Matter How You Look At It
One Zettabyte is indeed very, very big. Growing from just under 1 ZB to over 44 ZBs in a little over a decade is astounding. Managing this—whether you a technologist responsible for managing data, a business user who needs to use data, of a consumer just trying to manage the flood of all of your personal data—will be a challenge for all.