Microsoft Windows Server Benchmarked Against Six Linux Distributions

Written by Michael Larabel in Operating Systems on 23 January 2019 at 11:46 AM EST. Page 1 of 5. 20 Comments.

While it was not too long ago that Microsoft Windows Server 2019 began shipping and that we conducted some end-of-year benchmarks between Windows and Linux, with being in the process of running a number of Windows and Linux benchmarks as part of our ongoing 10GbE OS performance testing, I also took the opportunity to run some other benchmarks on Windows Server 2016 and 2019 as well as a set of Linux distributions.

With carrying out the fresh OS installations anyways for the network testing, with recently having brought over some more Phoronix Test Suite test profiles with Windows support, I decided to run some fresh Windows Server vs. Linux benchmarks anyways. Granted, not all of the tests are server-oriented and not all of the traditional Linux server distributions were used. Just take this as you wish of some fresh Windows vs. Linux performance benchmarks.

All of the tests were done on the Tyan GT24E-B7106 that we commonly use for our Intel Xeon Scalable testing. This server has dual Xeon Gold 6138 processors yielding a total of 40 cores / 80 threads, 12 x 8GB DDR4-2666 memory, and Samsung 970 EVO NVMe storage. Obviously the hardware was the same throughout benchmarking all of these operating systems and all in their stock configurations; any reported differences in the automated table come down to just how each OS was exposing all of its components.

The operating systems benchmarked as part of this round of testing were:

- Antergos 19.1
- Clear Linux 27400
- Debian 9.6
- openSUSE Tumbleweed
- Ubuntu 18.04 LTS
- Ubuntu 18.10
- Windows Server 2016
- Windows Server 2019

Each OS had all available software updates as of testing time. All of these Windows and Linux benchmarks were carried out using the open-source and fully-automated Phoronix Test Suite benchmarking software.


Related Articles