LZO

LZO, short for Lempel-Ziv-Oberhumer, is a family of lossless data compression algorithms and a free software library designed for high-speed operation and a small memory footprint. It was developed by Markus Oberhumer and colleagues as a fast alternative to other compression schemes, combining a LZ77-based core with specialized encoding for matches and literals. The LZO project provides multiple profiles, with the commonly used LZO1X and LZO2 variants, offering different balances of speed and compression ratio. The reference implementation is written in C and supports both in-memory and streaming compression and decompression, as well as small code size suitable for embedded environments.

LZO emphasizes speed over maximum compression, delivering fast real-time compression suitable for applications where latency matters. Data-dependent ratios vary, but LZO typically achieves modest to good compression (often outperforming many fast algorithms in speed while producing lower ratios than more aggressive methods). It is widely used in embedded systems, archival tools, and networking scenarios where throughput is critical. In the Linux ecosystem, LZO has been offered as a compression option for certain filesystems and packaging workflows, and the lzop command-line utility implements the LZO algorithm for practical use.

The LZO library is available as free software under a permissive license, encouraging integration into a variety of projects and devices. It has been ported to numerous platforms and languages, with various bindings and optimizations. LZO should be distinguished from similarly named tools that use the LZO core, such as the standalone lzop utility.