Throughout life, we always find something to measure: the amount of food we need to cook for the family, the length and width of that couch you want to put in the room, our weight and height. The latter is especially exciting: it’s really cool to learn that in just one year you grew by a full 2 inches!

Each measurement requires an instrument and its own unit of measurement. For example, body weight is measured with scales in kilograms (or pounds), time is measured with clocks in seconds, etc. But how does one measure information?

Bit: the smallest unit of information

The information entered into the computer should be specific and unambiguous. For a long time, people have used ciphers. The simplest and most convenient of them were digital. Any information from the names of flowers to the days of the week can be presented in the form of numbers. When processed with a conventional computer, the data is encoded by numbers. They are represented by the electrical signals that the computer works with.

For the convenience of distinguishing, signals of two levels are used in classical electronic computers. One of them corresponds to the number 1, and the other to 0. Any letter, sound, or image on a computer is represented by a set of numbers. The numbers 1 and 0 are called binary. These are the symbols that make up the language understood and used by the computer. Any information on the computer is represented by binary digits: 1, meaning “there is a signal” or “there is a high signal” and 0, meaning “no signal” or “there is a low signal”.

The smallest unit of information is the bit (b).

Each digit of the machine binary code carries the amount of information equal to one bit. It can take only one of two values: either 1 or 0. It is very inconvenient to measure information in bits because the numbers come out too big.

Byte: a sequence of eight bits

Since people do not consider the mass of ships in grams, larger and hence more convenient units were invented for measuring information as well.

The processing of information takes place in the processor. This is a device that can work with several bits at once (8, 16, 32, 64, …). The first computers processed 8 bits of information simultaneously, so we needed a new unit of measurement which was called a byte (B), which means 8 bits.

Bit marks are easily confused with byte marks. Note that the abbreviations for bit numbers use the lowercase letter “b”, while bytes are denoted using the capital “B”.

Large units of information

There are even larger units of information since modern computers process huge amounts of information significantly exceeding bytes.

The computer industry has historically used the units kilobytemegabyte, and gigabyte in at least two slightly different measurement systems, which are slightly contradictory to each other.

  • The first one is a decimal-based system, which uses bytes in the powers of ten: kilobyte (10^3 bytes), megabyte (10^6 bytes), gigabyte (10^9 bytes), and so on. These units are used by the International System of Units (SI).
  • The second one is a binary-based system that uses bytes in the powers of two: kibibyte (2^10 bytes), mebibyte (2^20 bytes), gibibyte (2^30 bytes), and so on. This system was actively used to describe computer memory.

To resolve this confusion, the International Electrotechnical Commission (IEC) suggested using prefixes kilomega, and giga only for the decimal-based system and using new prefixes kibimebi, and gibi for the binary-based system. Here bi means binary: kibibyte is kilo binary byte.

Here is a table with commonly used units of information according to modern international standards.

SI metricSymbolPowers of tenIEC metricSymbolPowers of two
KilobytekB10^3 B (1000 B)KibibyteKiB2^10 B (or 1024 B)
MegabyteMB10^6 B (1000 kB)MebibyteMiB2^20 B (or 1024 KiB)
GigabyteGB10^9 B (1000 MB)GibibyteGiB2^30 B (or 1024 MiB)
TerabyteTB10^12 B (1000 GB)TebibyteTiB2^40 B (or 1024 GiB)
PetabytePB10^15 B (1000 TB)PebibytePiB2^50 B (or 1024 TiB)

Of course, not all units of measurement are listed here. We hope this classification will not cause you any difficulties. It is good that a byte is always 8 bits. But even this did not come immediately.

Note that some people and organizations still prefer kilomega, and giga to describe powers of two. In this course, we follow the recommendations of the IEC and use modern prefixes kibimebi, and gibi.

Measurement units conversion

To strengthen your newly obtained knowledge, let’s look at the solution of a rather typical problem where you need to convert 1 GiB to KiB. When we convert bigger units into smaller ones, we need to resort to an arithmetic operation called multiplication:

1 GiB = 1KiB * 1024 * 1024 = 1048576 KiB

Accordingly, when you need to convert small units into big ones, you use division. Let’s try to convert 16384 bits to KiB:

16384 bits = (16384 / 8) / 1024 = 2 KiB

If you want to convert 1 GB to kB, you should multiply the number by a thousand twice:

1 GB = 1kB * 1000 * 1000 = 1000000 kB

Congratulations, now you have studied one of the basic topics of computer science and are ready to storm new dizzying heights of knowledge.


To wrap-up:

  • The smallest unit of information is a bit.
  • One byte consists of 8 bits.
  • For convenience in computer science, we use binary-based units of information as bytes, kibibytes, mebibytes, etc.
  • To convert one kibibyte to bytes, we need to multiply one by 2 ^ {10}, to convert one mebibyte, we need to multiply one by 2 ^ {20}, etc.

Leave a Reply