1LZ4 - Extremely fast compression 2================================ 3 4LZ4 is lossless compression algorithm, 5providing compression speed > 500 MB/s per core, 6scalable with multi-cores CPU. 7It features an extremely fast decoder, 8with speed in multiple GB/s per core, 9typically reaching RAM speed limits on multi-core systems. 10 11Speed can be tuned dynamically, selecting an "acceleration" factor 12which trades compression ratio for faster speed. 13On the other end, a high compression derivative, LZ4_HC, is also provided, 14trading CPU time for improved compression ratio. 15All versions feature the same decompression speed. 16 17LZ4 is also compatible with [dictionary compression](https://github.com/facebook/zstd#the-case-for-small-data-compression), 18both at [API](https://github.com/lz4/lz4/blob/v1.8.3/lib/lz4frame.h#L481) and [CLI](https://github.com/lz4/lz4/blob/v1.8.3/programs/lz4.1.md#operation-modifiers) levels. 19It can ingest any input file as dictionary, though only the final 64KB are used. 20This capability can be combined with the [Zstandard Dictionary Builder](https://github.com/facebook/zstd/blob/v1.3.5/programs/zstd.1.md#dictionary-builder), 21in order to drastically improve compression performance on small files. 22 23 24LZ4 library is provided as open-source software using BSD 2-Clause license. 25 26 27|Branch |Status | 28|------------|---------| 29|master | [![Build Status][travisMasterBadge]][travisLink] [![Build status][AppveyorMasterBadge]][AppveyorLink] [![coverity][coverBadge]][coverlink] | 30|dev | [![Build Status][travisDevBadge]][travisLink] [![Build status][AppveyorDevBadge]][AppveyorLink] | 31 32[travisMasterBadge]: https://travis-ci.org/lz4/lz4.svg?branch=master "Continuous Integration test suite" 33[travisDevBadge]: https://travis-ci.org/lz4/lz4.svg?branch=dev "Continuous Integration test suite" 34[travisLink]: https://travis-ci.org/lz4/lz4 35[AppveyorMasterBadge]: https://ci.appveyor.com/api/projects/status/github/lz4/lz4?branch=master&svg=true "Windows test suite" 36[AppveyorDevBadge]: https://ci.appveyor.com/api/projects/status/github/lz4/lz4?branch=dev&svg=true "Windows test suite" 37[AppveyorLink]: https://ci.appveyor.com/project/YannCollet/lz4-1lndh 38[coverBadge]: https://scan.coverity.com/projects/4735/badge.svg "Static code analysis of Master branch" 39[coverlink]: https://scan.coverity.com/projects/4735 40 41> **Branch Policy:** 42> - The "master" branch is considered stable, at all times. 43> - The "dev" branch is the one where all contributions must be merged 44 before being promoted to master. 45> + If you plan to propose a patch, please commit into the "dev" branch, 46 or its own feature branch. 47 Direct commit to "master" are not permitted. 48 49Benchmarks 50------------------------- 51 52The benchmark uses [lzbench], from @inikep 53compiled with GCC v8.2.0 on Linux 64-bits (Ubuntu 4.18.0-17). 54The reference system uses a Core i7-9700K CPU @ 4.9GHz (w/ turbo boost). 55Benchmark evaluates the compression of reference [Silesia Corpus] 56in single-thread mode. 57 58[lzbench]: https://github.com/inikep/lzbench 59[Silesia Corpus]: http://sun.aei.polsl.pl/~sdeor/index.php?page=silesia 60 61| Compressor | Ratio | Compression | Decompression | 62| ---------- | ----- | ----------- | ------------- | 63| memcpy | 1.000 | 13700 MB/s | 13700 MB/s | 64|**LZ4 default (v1.9.0)** |**2.101**| **780 MB/s**| **4970 MB/s** | 65| LZO 2.09 | 2.108 | 670 MB/s | 860 MB/s | 66| QuickLZ 1.5.0 | 2.238 | 575 MB/s | 780 MB/s | 67| Snappy 1.1.4 | 2.091 | 565 MB/s | 1950 MB/s | 68| [Zstandard] 1.4.0 -1 | 2.883 | 515 MB/s | 1380 MB/s | 69| LZF v3.6 | 2.073 | 415 MB/s | 910 MB/s | 70| [zlib] deflate 1.2.11 -1| 2.730 | 100 MB/s | 415 MB/s | 71|**LZ4 HC -9 (v1.9.0)** |**2.721**| 41 MB/s | **4900 MB/s** | 72| [zlib] deflate 1.2.11 -6| 3.099 | 36 MB/s | 445 MB/s | 73 74[zlib]: http://www.zlib.net/ 75[Zstandard]: http://www.zstd.net/ 76 77LZ4 is also compatible and optimized for x32 mode, 78for which it provides additional speed performance. 79 80 81Installation 82------------------------- 83 84``` 85make 86make install # this command may require root permissions 87``` 88 89LZ4's `Makefile` supports standard [Makefile conventions], 90including [staged installs], [redirection], or [command redefinition]. 91It is compatible with parallel builds (`-j#`). 92 93[Makefile conventions]: https://www.gnu.org/prep/standards/html_node/Makefile-Conventions.html 94[staged installs]: https://www.gnu.org/prep/standards/html_node/DESTDIR.html 95[redirection]: https://www.gnu.org/prep/standards/html_node/Directory-Variables.html 96[command redefinition]: https://www.gnu.org/prep/standards/html_node/Utilities-in-Makefiles.html 97 98 99Documentation 100------------------------- 101 102The raw LZ4 block compression format is detailed within [lz4_Block_format]. 103 104Arbitrarily long files or data streams are compressed using multiple blocks, 105for streaming requirements. These blocks are organized into a frame, 106defined into [lz4_Frame_format]. 107Interoperable versions of LZ4 must also respect the frame format. 108 109[lz4_Block_format]: doc/lz4_Block_format.md 110[lz4_Frame_format]: doc/lz4_Frame_format.md 111 112 113Other source versions 114------------------------- 115 116Beyond the C reference source, 117many contributors have created versions of lz4 in multiple languages 118(Java, C#, Python, Perl, Ruby, etc.). 119A list of known source ports is maintained on the [LZ4 Homepage]. 120 121[LZ4 Homepage]: http://www.lz4.org 122