Latest update for 'fast' compression algorithm LZ4 sprints past old versions

The latest version of the LZ4 compression algorithm, version 1.10, has been released with significant performance improvements, achieving speeds nearly ten times faster than previous versions. This update utilizes multithreading to enhance compression speed, particularly on systems with multiple CPU cores. For instance, on an x64 Linux platform, decompression of a 5GB text file has improved from 5 seconds with version 1.9.4 to just 3 seconds with version 1.10, marking a performance increase of over 60%. LZ4 is known for its speed-optimized capabilities and has been integrated into the Linux kernel since version 3.11. The algorithm's enhancements are attributed to the contributions of Takayuki Matsuoka, who previously developed a parallelized version called lz4mt. The new version allows for more efficient use of CPU cores, with tests showing substantial speed increases on both AMD and Intel processors. This release highlights the challenges of writing efficient parallel code, as many applications still do not fully utilize multiple cores. The improvements in LZ4 1.10 are expected to benefit various Linux applications, particularly those that rely heavily on swap memory, as it will enhance the performance of the zswap feature in the Linux kernel. Overall, the advancements in LZ4 demonstrate the ongoing evolution of compression algorithms to meet the demands of modern computing environments.