Benchmarking Performance of a Virtual Machine
I have run a number of benchmarks to document the performance of Windows 10 running as a virtual machine on Linux, in the hope other PC users will dive into the fascinating world of virtualization (VFIO).
Benchmarks are helpful in comparing one system with another, and one configuration with another. I use them to optimize my Windows 10 performance and to make sure that updates/upgrades haven’t produced unwanted side effects.
Obviously the most important benchmark is the user experience. If you feel that something is wrong, sluggish, etc., then all the benchmarks in the world can’t help achieve a different experience.
Warning: If you benchmark, don’t overdo it! Particularly disk benchmarks have the unwanted side effect of reducing the lifetime of an SSD.
SteamVR Performance Test
My latest benchmark is the SteamVR Performance Test. This benchmark runs on a Windows 10 (1803) VM and tests if the machine is ready for VR (virtual reality).
The one thing left to do is to carry my PC to the next room and connect it to an actual HTC Vive. If time permits, I’ll do that.
Passmark 8 Benchmark
I’m using Passmark 8 instead of the latest version 9 since that is the version I bought. Passmark has a nice breakdown of different benchmarks, including 2D graphics performance tests, which are important to what I do under Windows. The benchmarks run on a Windows 10 (1803) VM, and I have included the benchmarks of a Windows 10 (1803) bare metal installation for comparison.
Legend:
- Red graph – Windows 10 (1803) on bare metal
- Blue graph – Windows 10 (1803) VM with Spectre protection enabled
- Yellow graph – Windows 10 (1803) VM with Spectre protection disabled
Here is the Passmark Rating for my Windows 10 Virtual Machine:
