Often described as “the new oil”, data has quickly become the most valuable asset to the business world. Organisations now collect data in amounts that were considered unimaginable even a decade ago, and the upwards trend is showing no signs of slowing. These enormous quantities of data need to be estimated in some way, which is why measurement concepts that might have seemed purely hypothetical not long ago are now becoming part of our daily jargon.
Such was the case with terabytes. When the first 1TB hard drive, manufactured by Hitachi, was launched in 2007, the achievement was immortalised in the Guinness World Records. However, 15 years later, a 1TB portable drive isn’t considered that exceptional at all – that is the nature of progress.
The next step up from a terabyte is petabyte (PB), which is a term used to describe one quadrillion bytes. ‘Quadrillion’ is yet another descriptor which seems to be only used in purely theoretical circumstances, yet in context, is used to signify 1,000,000 billions.
So why would anyone need one quadrillion bytes? Probably not the average person, that is why we won’t be seeing 1PB personal storage drives anytime soon. For the time being, petabytes are usually used to estimate the data held in enterprise-grade infrastructure, or storage modules for large data centres. Customers of such usually include research institutions and universities, as well as some of the largest financial and pharmaceutical organisations.
However, even if you’re not currently gearing up to hold petabytes of data, it’s still worth acquainting yourself with the idea. After all, even yotta- and zettabytes might not be too far in the future.
Mega, giga, tera... what's the difference?
Every measurement of storage is made up of bytes and each byte is made up of 8 bits, right at the bottom of the storage pile. From byte, we move up the size ladder, with kilobyte next in line, followed by megabyte, gigabyte, terabyte, petabyte, exabyte, and the unfathomably large zettabyte and yottabyte.
Establishing a strong foundation for DataOps
How to gain a competitive advantage with your available data
In order to convert from a lower metric to a higher, the rule is to multiply the number by 1,024 - so 1 terabyte is the same as 1,024 gigabytes.
So for example: 1 GB = 1,024 MB = 1,048,576 KB = 1,073,741,824 B.
That may sound strange if you've always been told that a kilobyte was a thousand bytes, and a megabyte a thousand kilobytes. The confusion is in part due to the use of binary in computers instead of decimal - 1,000 would be 1111101000 in binary, which isn't a convenient group size to use for a storage metric.
The use of 'kilo' is also confusing, given that a kilometre is 1,000 metres, not 1,024. In fact, in 1998 the International Electrotechnical Commission introduced new standardised prefixes, including "kibibytes" instead of kilobytes and "mebibytes" instead of megabytes... but they never quite caught on.
Here's a comparison between the various sizes:
| Metric | Exponent | Size (bytes) |
| Byte (B) | 1 | 1 |
| Kilobyte (KB) | 1024^1 | 1,024 |
| Megabyte (MB) | 1024^2 | 1,048,576 |
| Gigabyte (GB) | 1024^3 | 1,073,741,824 |
| Terabyte (TB) | 1024^4 | 1,099,511,627,776 |
| Petabyte (PB) | 1024^5 | 1,125,899,906,842,624 |
| Exabyte (EB) | 1024^6 | 1,152,921,504,606,846,976 |
| Zettabyte (ZB) | 1024^7 | 1,180,591,620,717,411,303,424 |
| Yottabyte (YB) | 1024^8 | 1,208,925,819,614,629,174,706,176 |
