What is TAR?

What is TAR?

TAR (Tape Archive) is a file packaging format originating from Unix in the late 1970s. The name refers to its original purpose: archiving files to magnetic tape drives for backup. TAR bundles multiple files and directories into a single archive file while preserving file system metadata including file permissions, ownership, timestamps, hard links, and symbolic links - attributes that are essential in Unix and Linux environments but which ZIP archives do not reliably preserve.

A plain TAR file does not compress data - it only bundles files together, similar to how you might rubber-band a stack of papers. Compression is applied on top of the TAR archive using separate tools: gzip creates .tar.gz, bzip2 creates .tar.bz2, and xz creates .tar.xz. This separation of concerns (packaging vs. compression) is a Unix design philosophy. TAR archives are the universal distribution format for Linux software source code and system backups.

How TAR Works

The name tar is short for tape archive, and that origin explains everything about the format. A .tar file bundles many files and directories into one continuous stream so they can be written to a backup tape, or today a single disk file, in one pass. The crucial point is that tar does no compression at all: a tarball is roughly the same size as the sum of its contents. Its job is packaging, not shrinking.

The format is deliberately simple. Each file in the archive is preceded by a fixed-size header block of 512 bytes containing the filename, size, permissions, owner and group, and modification time, followed by the file's raw contents padded up to a multiple of 512 bytes. The archive is just these header-and-data pairs laid end to end, with a run of zero blocks marking the end. Because it stores full Unix metadata, tar faithfully preserves permissions, symlinks, and ownership, which is exactly why it remains the backbone of Unix backups.

That streaming design is tar's superpower and its limitation. Because there is no central index, tar can create or extract an archive while reading a single continuous flow of bytes, which is ideal for piping. The flip side is that finding one file deep in a large tarball means scanning from the start, since there is no directory to jump to.

To get a compressed archive you pipe tar's output through a compressor. This is why the compound extensions .tar.gz, .tar.bz2, and .tar.xz are so common: tar builds the single stream and gzip, bzip2, or xz squeezes it. The famous tar czf archive.tar.gz files/ command does both steps at once, the z invoking gzip.

History and Standardization

tar dates back to the 1970s Unix era, when magnetic tape was the primary backup medium and an archive naturally meant a serial stream. The format long outlived the hardware it was named for, becoming the default way to move collections of files around the Unix world.

Over the decades several header variants appeared, and the format was standardized as part of POSIX. The widely used ustar format and the later pax extended format resolved early limits on filename length and file size, and modern tar implementations such as GNU tar and bsdtar interoperate around these agreed layouts.

Because it is simple, stable, and openly specified, tar support is universal on Unix-like systems and is built into most cross-platform archive tools. It underpins software distribution, container image layers, and backup systems, and its role as the packaging half of the tar-plus-compressor pairing keeps it in constant daily use.

TAR Compared to ZIP and 7z

The biggest conceptual difference from ZIP is division of labour. ZIP both archives and compresses each file individually inside one container, and it keeps a central directory for fast random access. tar only archives, leaving compression to a separate tool, and has no index. This makes tar better at streaming and at preserving Unix metadata, while ZIP is better for quickly extracting a single file from a large archive.

tar's separation of jobs gives it one real advantage over ZIP's per-file compression: because a compressor sees the whole concatenated stream at once, it can exploit redundancy across many small similar files, so a .tar.gz of a source tree often ends up smaller than the equivalent ZIP. The cost is that you cannot extract one file without decompressing up to its position.

Against 7z, tar again differs by philosophy: 7z is an all-in-one archiver with strong LZMA compression and an index, whereas tar leans on the Unix toolchain. On Windows, ZIP and 7z dominate; on Linux and macOS, tar plus a compressor remains the idiomatic choice, especially where preserving exact permissions and ownership matters.

Technical Details

TAR vs Other Archive Formats

TAR only bundles files and is usually paired with gzip or xz, whereas ZIP bundles and compresses in a single step.

Pros and Cons

When to Use TAR

Here are the most common situations where TAR is the right choice: