OpenSUSE Leap 15.1 Beta Is Running Well - Benchmarks On AMD EPYC Workstation

With openSUSE Leap 15.1 reaching beta this week I decided to take it for a quick spin of this Linux distribution derived from the same sources as SUSE Linux Enterprise 15 SP1. Here are some quick benchmarks compared to Leap 15.0 as well as the latest rolling-release openSUSE Tumbleweed.

OpenSUSE Leap 15.1 remains under active development and is expected to be officially released in May. But given reaching the beta state and being curious how the performance has evolved compared to openSUSE Tumbleweed, I ran some initial benchmarks of this beta snapshot this week. This preliminary round of tests was done using an AMD EPYC 7351P workstation with ASRockRack EPYCD8-2T, 8 x 4GB DDR4-2666 memory, and 800GB Intel DC P3600 (SSDPE2ME800G4) NVMe solid-state drive. Tests on more hardware will come as the openSUSE Leap 15.1 stable release approaches.

OpenSUSE Leap 15.1 Beta remains on the Linux 4.12 kernel albeit with its share of back-ports and extra patches added atop as is common for the enterprise Linux distributions. OpenSUSE Leap 15.1 is using the GCC 7.4.0 compiler compared to GCC 7.3.1 on Leap 15.0. For those using openSUSE Leap on the desktop, 15.1 is currently using KDE Plasma 5.12.7 compared to 5.12.6 on 15.0. Another change on the desktop side is 15.1 moving up to the X.Org Server 1.20 series where as 15.0 shipped with xorg-server 1.19.6.

Meanwhile, the rolling-release openSUSE Tumbleweed is currently riding with the Linux 4.20 kernel, GCC 8.2.1 compiler, and KDE Plasma 5.15 desktop with X.Org Server 1.20. Another change worth pointing out form the performance perspective is Tumbleweed defaulting to MQ-Deadline for NVMe SSD storage while openSUSE Leap 15.1 uses no I/O scheduler in such configuration. All three openSUSE installations tested were installed cleanly with their default partitioning layout on this AMD EPYC 7351P workstation.

Here's a look at various benchmarks via the Phoronix Test Suite.