A Look At The Clear Linux Performance Over The Course Of 2018
With the end of the year quickly approaching, it's time for our annual look at how the Linux performance has evolved over the past year from graphics drivers to distributions. This year was a particularly volatile year for Linux performance due to Spectre and Meltdown mitigations, some of which have at least partially recovered thanks to continued optimizations landing in subsequent kernel releases. But on the plus side, new releases of Python, PHP, GCC 8, and other new software releases have helped out the performance. For kicking off our year-end benchmark comparisons, first up is a look at how Intel's performance-optimized Clear Linux distribution evolved this year.
For getting a look at the performance, on four different systems (two Xeon boxes, a Core i5, and Core i7 systems), the performance was compared from Clear Linux at the end of 2017 to the current rolling-release state as of this week.
Clear Linux at the end of 2017 was shipping with the Linux 4.14 kernel, GCC 7.2.1 compiler, EXT4 file-system,no I/O scheduler for NVMe storage and Kyber for SATA SSDs, no Spectre/Meltdown mitigations, and the other latest releases as of that time.
Clear Linux as of its recent 26760 build now has the Linux 4.19 kernel with all relevant Spectre/Meltdown mitigations, GCC 8.2.1, EXT4 file-system, Python 3.7, is defaulting to the MQ-DEADLINE I/O scheduler, and the other bleeding-edge software packages.
The boxes used for testing were the Xeon E3-1235L v5, Xeon E5-2609 v4, Core i5 6500, and Core i7 7700K. These benchmarks were carried out using the Phoronix Test Suite.
The SQLite performance for a basic I/O test saw roughly the same performance between last year and now. There were some more sizable performance hits towards the start of 2018 due to KPTI and Retpolines, but that has improved a lot with time where as of now the Xeon E3-1235L v5 box is slightly faster while the three other boxes were slightly slower. There's also been a lot of EXT4 file-system activity this year.
The IOzone benchmark also shows a small hit on three of the systems with the current state of 2018, but overall not that bad as some may have feared at the start of 2018.
Moving onto compute/system tests, the FFTE performance improved faintly over the course of this year, possibly due to the GCC8 compiler upgrade.
The SciMark2 synthetic test largely impacted by compiler changes saw similar performance.