Datachunk

Datachunk, or data chunk, is a term used in computing to describe a contiguous block of data that constitutes a discrete unit within a larger container, file format, or data stream. A data chunk has a defined start and end and is often accompanied by metadata such as an identifier and length.

In file formats that use a chunked layout, a data chunk carries the actual payload for a specific media or information type. For example, the RIFF family used by WAV audio and many video containers stores chunks as a four-character ID, a 32-bit size, and a payload. The "data" chunk in WAV files holds the raw sample data. Most RIFF-based containers pad chunks to even sizes.

AIFF and other formats also use a chunked structure, with designated chunks that contain the media data; endianness and field sizes vary by format.

In data transmission and streaming, the term data chunk describes a segment of a message or body that is transferred as a unit. Protocols may prefix chunks with length and type information, enabling parsers to locate bounds, validate integrity, and support streaming or error recovery.

In data processing and databases, a data chunk can refer to a fixed-size block read or written as a unit, aiding buffering and parallel processing.

For parsing, software typically searches for known chunk identifiers, checks the reported size against bounds, and gracefully ignores unknown chunks to maintain compatibility.

See also: chunk, RIFF, WAV, AIFF, HTTP chunked transfer encoding.