Clear Linux Concluding 2019 With ~7% Faster Performance For The Year, Some Open-Source Workloads Much Faster
Written by Michael Larabel in Operating Systems on 17 December 2019 at 09:47 AM EST. Page 3 of 6. 5 Comments.
In some of the Intel MKL-DNN / DNNL (Deep Neural Network Library) tests were the most significant improvements when moving from Clear Linux at the end of 2018 to EOY 2019. Many of the DNNL improvements were when running with u8s8f32 data types for AVX-512 over f32.
Video encoding benchmarks built on both Clear Linux releases were also seeing some improvements, including for Intel's own Scalable Video Technology encoders.
Single-threaded audio encoding also came out ahead.