Binary prefix

A binary prefix is a unit prefix for multiples of units in data processing, data transmission, and digital information, most notably the bit and the byte, to indicate multiplication by a power of 2. These prefixes are primarily used in the computer industry to quantify digital data storage and memory capacity, providing a standardized method to distinguish base-2 multiples (powers of 1024) from the base-10 multiples (powers of 1000) denoted by standard International System of Units (SI) prefixes.

History and background

In the early days of computing, computer scientists and engineers needed a convenient way to express large quantities of digital data. Because computer memory and storage architectures are inherently binary, capacities naturally scale in powers of 2. The value of $2^{10}$ (1,024) is very close to $10^3$ (1,000), which led the industry to adopt the SI prefix "kilo" to represent 1,024 bytes. Similarly, "mega" was used for $2^{20}$ (1,048,576) bytes, and "giga" for $2^{30}$ (1,073,741,824) bytes.

While this shorthand was convenient, it created significant ambiguity. The SI prefixes strictly denote powers of 10 in all other scientific and engineering disciplines. As storage capacities grew, the discrepancy between the binary and decimal interpretations became more pronounced. Furthermore, hard drive manufacturers began using the strict decimal definitions (e.g., 1 megabyte = 1,000,000 bytes) to advertise larger numerical capacities, while operating systems continued to calculate and display storage using the binary definitions. This discrepancy led to widespread consumer confusion and even class-action lawsuits against storage manufacturers.

IEC standard prefixes

To resolve this ambiguity, the International Electrotechnical Commission (IEC) introduced a new set of specific binary prefixes in 1998, which were later incorporated into the IEC 60027-2 standard. These prefixes were designed to be used exclusively for powers of 1024, leaving the SI prefixes to their strict decimal meanings.

The IEC binary prefixes are formed by taking the first two letters of the corresponding SI prefix and adding "bi" (for binary). The symbols are formed by taking the SI symbol and adding an "i". The standard IEC binary prefixes are:

• Kibi (Ki): $2^{10}$ = 1,024
• Mebi (Mi): $2^{20}$ = 1,048,576
• Gibi (Gi): $2^{30}$ = 1,073,741,824
• Tebi (Ti): $2^{40}$ = 1,099,511,627,776
• Pebi (Pi): $2^{50}$ = 1,125,899,906,842,624
• Exbi (Ei): $2^{60}$ = 1,152,921,504,606,846,976

Under this system, a kilobyte (kB) is strictly 1,000 bytes, while a kibibyte (KiB) is strictly 1,024 bytes.

Divergence between decimal and binary multiples

The numerical difference between the decimal (SI) and binary (IEC) interpretations increases as the magnitude of the prefix grows. While the difference at the "kilo/kibi" level is relatively small, it becomes highly significant at higher capacities:

• Kilo / Kibi: 1 kB = 1,000 B; 1 KiB = 1,024 B (2.4% difference)
• Mega / Mebi: 1 MB = 1,000,000 B; 1 MiB = 1,048,576 B (4.9% difference)
• Giga / Gibi: 1 GB = 1,000,000,000 B; 1 GiB = 1,073,741,824 B (7.4% difference)
• Tera / Tebi: 1 TB = 1,000,000,000,000 B; 1 TiB = 1,099,511,627,776 B (10.0% difference)
• Peta / Pebi: 1 PB = $10^{15}$ B; 1 PiB = $2^{50}$ B (12.6% difference)
• Exa / Exbi: 1 EB = $10^{18}$ B; 1 EiB = $2^{60}$ B (15.3% difference)

Adoption and industry usage

The adoption of IEC binary prefixes has been mixed across the technology industry, largely depending on the specific hardware or software context.

Computer Memory (RAM): Solid-state memory, such as RAM, is manufactured in capacities that are strict powers of 2. The JEDEC Solid State Technology Association continues to use the traditional terms (kilobyte, megabyte, gigabyte) to denote binary multiples in memory specifications, though IEC prefixes are increasingly recognized in technical documentation.

Storage Devices: Manufacturers of hard disk drives (HDDs), solid-state drives (SSDs), and flash media universally use decimal SI prefixes. A "1 Terabyte" hard drive contains exactly $10^{12}$ bytes.

Operating Systems: The way operating systems report file sizes and storage capacity varies. Microsoft Windows continues to use the traditional SI abbreviations (KB, MB, GB) but calculates them using binary multiples (base-1024), effectively displaying GiB but labeling it as GB. Apple's macOS transitioned to strict decimal SI prefixes (base-1000) starting with Mac OS X Snow Leopard (10.6) to align with hardware manufacturer advertising. Many Linux distributions and utilities (such as the GNU coreutils) have adopted the IEC standard, using KiB, MiB, and GiB for base-1024 calculations, and kB, MB, and GB for base-1000.

Data Transmission: In networking and telecommunications, data transfer rates (e.g., megabits per second, Mbps) are universally measured using decimal SI prefixes. A 100 Mbps network connection transfers 100,000,000 bits per second, not 104,857,600.

Standardization and formal recognition

The IEC binary prefixes have been formally adopted and endorsed by several major international standards organizations. In 2002, the Institute of Electrical and Electronics Engineers (IEEE) published IEEE 1541-2002, which officially recommended the use of IEC prefixes for binary multiples. Furthermore, the International Organization for Standardization (ISO) incorporated the IEC prefixes into the ISO/IEC 80000-13:2008 standard (Quantities and units — Part 13: Information science and technology), cementing their status as the internationally recognized standard for denoting binary multiples in information technology. Despite these formal standardizations, colloquial use of SI prefixes to mean binary multiples remains deeply entrenched in everyday consumer and professional vocabulary.