Isma university

3. Units of information

In computer science, as a rule, information represented by a discrete signal is measured. In this case, the following approaches are distinguished:
1. structural. Measures the amount of information by simply counting the information items that make up a message. It is used to assess the capabilities of storage devices, the volume of transmitted messages, coding tools without taking into account the statistical characteristics of their operation.
2. statistical. Takes into account the likelihood of messages: the message that is less likely is considered more informative, i.e., least expected. It is used to assess the significance of the information received.
3.semantic. Takes into account the appropriateness and usefulness of information. It is used to assess the effectiveness of the information received and its compliance with reality.
Within the framework of the structural approach, three measures of information are distinguished:
• geometric. Determines the maximum possible amount of information in a given volume. The measure can be used to determine the information storage capacity of a computer;
• combinatorial. Evaluates the possibility of presenting information using various combinations of information elements in a given volume. A combinatorial measure can be used to assess the information capabilities of a certain coding system;
• additive, or Hartley measure
Information units are used to measure the amount of information - a value calculated logarithmically. This means that when several objects are considered as one, the number of possible states is multiplied, and the amount of information is added. Most often, the measurement of information concerns the volume of computer memory and the volume of data transmitted via digital communication channels. Amounts of information can be represented as the logarithm of the number of states. The smallest integer whose logarithm is positive is 2. The corresponding unit, the bit, is the basis for calculating information in digital technology.
The unit corresponding to the number 3 (trit) is equal to a bit, the number 10 (hartley) is equal to a bit. A unit such as nat (nat), corresponding to the natural logarithm, is used in computing in engineering and scientific calculations. The base of natural logarithms is not an integer.
Integer numbers of bits correspond to the number of states equal to powers of two. A special name has 4 bits - nibble (nibble, tetrad, four binary digits), which contain the amount of information contained in one hexadecimal digit. The next most popular unit of information is 8 bits, or bytes. It is to the byte (and not to the bit) that all large amounts of information calculated in computer technologies are directly reduced.
Quantities such as a machine word, etc., which are several bytes, are almost never used as units.
To measure large numbers of bytes, use the units "kilobytes" = 1000 bytes and "KB" (kibibyte, kibibyte) = 1024 bytes. The units "megabyte" = 1000 kilobytes = 1,000,000 bytes and "MB" (mebibyte, mebibyte) = 1024 kb = 1,048,576 bytes are used to measure the volume of information carriers. The units “gigabyte” = 1000 megabytes = 1,000,000,000 bytes and “gigabytes” (gibibytes, gibibyte) = 1024 MB = 230 bytes measure the size of large storage media, such as hard drives. The difference between binary and decimal units is already over 7%. The size of the 32-bit address space is 4 GB ≈ 4.295 MB. The same order is for DVD-ROM and modern flash media. The sizes of hard drives are already reaching hundreds and thousands of gigabytes. To calculate even larger amounts of information, there are units of terabytes - tebibytes (1012 and 240, respectively), petabytes - pebibytes (1015 and 250, respectively), etc. [one; from. 115].
A byte is defined for a specific computer as the minimum step of memory addressing, which on older machines was not necessarily 8 bits. In modern tradition, a byte is often considered equal to eight bits. In such designations as byte (Russian) or B (English), byte (B) means exactly 8 bits, although the term "byte" is not entirely correct from the point of view of theory. For a long time, they tried not to attach much importance to the difference between the multipliers 1000 and 1024. To avoid misunderstandings, you should clearly understand the difference between: binary multiple units, designated according to GOST 8.417-2002 as "KB", "MB", "GB", etc. (two in powers of ten); units of kilobytes, megabytes, gigabytes, etc., understood as scientific terms (ten in powers of multiples of three). The latter, by definition, are respectively bytes. As terms for "KB", "MB", "GB", etc. IEC suggests “kibibyte”, “mebibyte”, “gibibyte”, etc., however, these terms are criticized for being unpronounceable and are not spoken about.