LZ4 1.11 Introduces Multi-Threaded Compression For liblz4, Other Performance Gains

For those making use of the LZ4 extremely fast compression algorithm, LZ4 1.11 released today and brings multi-threaded compression support to its library (liblz4). LZ4 1.10 introduced initial multi-threading support into the lz4 command line tool. With LZ4 1.11, multi-threading has been integrated into the liblz4 library so that applications relying on this library can also benefit from the better performance on today's multi-core CPUs. A new field allows controlling the number of workers/threads when leveraging liblz4. The results are promising from the project's testing on an AMD EPYC server: The lz4 command line tool also has a --memory option now available for keeping the multi-threaded compression within a memory budget. LZ4 1.11 also brings some nice improvements in general over the LZ4 1.10 release: Those interested in today's LZ4 1.11 update can learn more at GitHub.