LZ4 v1.9.4 Achieves 20~70% Speedups For Some CPUs & Configurations

LZ4 v1.9.4 is out today as the first point release in nearly two years for this BSD-licensed, speedy, lossless compression algorithm.

Yann Collet announced the release today of LZ4 v1.9.4 as while it's a "maintenance" release, there are around 350 patches and a number of improvements -- including performance optimizations. LZ4 v1.9.4 brings a ~20% speedup for data decompression on ARM64 platforms. The double-digit decompression improvement can be seen on Apple M1/M2 devices, modern ARM64 servers, and other newer 64-bit Arm SoCs.

In addition to the big ARM64 speed-up for decompression, there can be around a 70% improvement for some specific scenarios of data compression with small blocks and other factors involved. Rounding out the decompression speed-ups in today's release is around a 40% improvement for compressed data making use of the lz4frame format as the checksum validation during decompression can now be ignored.

LZ4 v1.9.4 also adds new capabilities to the LZ4 library, Makefile build fixes and updates, documentation improvements, and more. There is even Solaris 10 compatibility added in 2022 for the LZ4 build system code.

Downloads and more details on the LZ4 v1.9.4 release via GitHub.

For those interested I do have LZ4 compression/decompression benchmarks and will be moving to the v1.9.4 release shortly for some CPU benchmarks.