TAR.GZ isn’t a format with its own specification so much as a convention: pipe TAR’s 1979 file-bundling output through gzip’s 1992 DEFLATE compression, and save the result as one file, commonly nicknamed a “tarball.” The combination filled a gap neither tool solved alone \u2014 TAR bundles files with Unix metadata but doesn’t compress, gzip compresses a single stream but can’t bundle multiple files \u2014 and became the de facto standard for distributing open-source software source code and Unix system backups almost as soon as gzip existed as a patent-free alternative to the older compress utility.<\/p>
For a DAM, TAR.GZ is the natural format for anything moving between Unix-based systems: nightly backups of an asset library, migrating storage between servers, or packaging a large batch of files for a Linux-based downstream pipeline, all benefiting from TAR’s faithful metadata preservation and gzip’s broad, patent-free compatibility. The trade-off worth knowing is that because the whole archive is compressed as a single stream rather than per-file, pulling one asset out of a large .tar.gz means decompressing everything before it in the stream \u2014 which is fine for a backend restore job but a poor fit for a DAM feature like previewing one file inside an archive without unpacking the whole thing.<\/p>","protected":false},"excerpt":{"rendered":"
The Unix-standard two-stage archive: files are first bundled uncompressed with TAR, then the entire resulting stream is compressed with gzip, producing one compact, single-file package commonly called a tarball.<\/p>\n","protected":false},"author":0,"featured_media":0,"template":"","meta":{"footnotes":"","faq":[{"question":"Is .tar.gz its own file format or a combination of two?","answer":"It's a combination \u2014 TAR bundles multiple files (with metadata) into one uncompressed stream, and gzip then compresses that entire stream, so a .tar.gz is really 'a gzip file whose contents happen to be a tar archive.'"},{"question":"Why is .tar.gz sometimes written as .tgz?","answer":"They're the same thing \u2014 .tgz is a shortened alias used mainly on older filesystems that limited extensions to three characters, and both extensions are handled identically by extraction tools."},{"question":"Can you preview one file inside a .tar.gz without extracting the whole archive?","answer":"Not efficiently \u2014 because gzip compresses the whole TAR stream as one unit rather than per-file like ZIP, reaching any single file means decompressing everything before it in the stream first."},{"question":"Why do open-source projects usually distribute source code as .tar.gz instead of ZIP?","answer":"It's a long-standing Unix\/Linux convention: TAR preserves Unix file permissions and structure exactly, and gzip has been a patent-free, universally available compressor on Unix systems since the early 1990s, making .tar.gz the natural default for that ecosystem."},{"question":"Does a DAM need special tooling to handle .tar.gz files?","answer":"Not really \u2014 standard libraries in every server-side language (Python's tarfile, Node's tar package) read and write .tar.gz directly, and the built-in tar command on Linux\/macOS handles it with a single flag (-z)."},{"question":"Is .tar.gz better than ZIP for compression ratio?","answer":"Often yes for many similar or text-heavy files, since gzip compresses the whole bundled stream together and can exploit redundancy across files, not just within each one the way ZIP's per-file compression does."},{"question":"Can Windows extract .tar.gz files without third-party software?","answer":"Only via the command line \u2014 Windows' built-in tar.exe (added in Windows 10) can handle it from PowerShell\/cmd, but there's no native graphical double-click extraction the way ZIP has, so most users still install 7-Zip or WinRAR."},{"question":"When would a DAM choose .tar.gz over plain .zip for a backup?","answer":"When the source and destination are both Unix-like systems and preserving exact file permissions\/ownership matters, and when the collection includes many similar files where gzip's whole-stream compression yields a meaningfully smaller archive than ZIP's per-file approach."}],"checked_date":"2026-08-07","sources":[{"statement":"The tar utility dates to Seventh Edition Unix in 1979, and gzip was released by the GNU Project in 1992; combining them via a pipe to compress a tar stream became the standard '.tar.gz' convention on Unix systems.","source_name":"Wikipedia \u2014 Gzip","url":"https:\/\/en.wikipedia.org\/wiki\/Gzip","checked":"2026-08-07"},{"statement":"gzip compresses a single stream and has no archiving capability of its own, which is why it is conventionally combined with tar to bundle multiple files before compression.","source_name":"GNU Gzip Manual","url":"https:\/\/www.gnu.org\/software\/gzip\/manual\/gzip.html","checked":"2026-08-07"}],"kicker":"","fact_checker":0,"reading_time":0,"revisions":[],"seo_title":"","seo_description":"","noindex":false,"ext":"tar.gz","full_name":"Gzip-compressed Tape Archive (Tarball)","vendor_name":"AT&T (TAR, 1979) + GNU Project (gzip, 1992)","introduced":"TAR: 1979; gzip: 1992 \u2014 the combined .tar.gz convention followed shortly after gzip's release","container_notes":"A .tar.gz file is literally a gzip-compressed TAR stream \u2014 TAR contributes the file bundling and metadata (names, permissions, ownership) with no compression of its own, while gzip's DEFLATE algorithm compresses that entire concatenated stream as one unit rather than compressing each file separately the way ZIP does. That single-stream compression usually yields a smaller result than ZIP for many similar files (since redundancy across files, not just within one, gets compressed away), but it also means extracting or even listing one file requires decompressing everything up to that point in the stream.","open_with":[{"platform":"Linux\/Unix\/macOS","how":"tar -xzf archive.tar.gz decompresses and extracts in one command using the built-in tar tool"},{"platform":"Windows","how":"7-Zip or WinRAR extract .tar.gz in two visible steps (first to .tar, then to files) or in one pass depending on the tool"},{"platform":"GUI archive managers","how":"Archive Manager\/File Roller (Linux) and Keka (macOS) handle .tar.gz through a standard double-click interface"},{"platform":"DAM\/asset pipeline","how":"Python's tarfile module (mode 'r:gz') or Node's tar package read and write .tar.gz programmatically in one pass, standard for automated backup and open-source software distribution pipelines"}],"archive_advice":"TAR.GZ is the default, well-understood choice for a DAM's server-side backups and Unix-to-Unix asset transfers, since it preserves file metadata faithfully and compresses efficiently with tooling built into every Linux distribution. For delivering assets to end users who may be on Windows without extra software installed, ZIP remains the safer default, since .tar.gz extraction is less universally native outside Unix-like systems.","related":[],"tool_link":""},"silo":[23],"class_list":["post-3048","format","type-format","status-publish","hentry","silo-formats"],"_links":{"self":[{"href":"https:\/\/picajet.com\/articles\/wp-json\/wp\/v2\/format\/3048","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/picajet.com\/articles\/wp-json\/wp\/v2\/format"}],"about":[{"href":"https:\/\/picajet.com\/articles\/wp-json\/wp\/v2\/types\/format"}],"version-history":[{"count":0,"href":"https:\/\/picajet.com\/articles\/wp-json\/wp\/v2\/format\/3048\/revisions"}],"wp:attachment":[{"href":"https:\/\/picajet.com\/articles\/wp-json\/wp\/v2\/media?parent=3048"}],"wp:term":[{"taxonomy":"silo","embeddable":true,"href":"https:\/\/picajet.com\/articles\/wp-json\/wp\/v2\/silo?post=3048"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}