New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Common methods to benchmark a Linux VPS?
jungleparrot
Member
in Help
I have a few VPS's at different providers and was wanting to compare them for things like network speed and disk read/write. It's there a common script or package that can be used for collecting that data?
Comments
https://github.com/masonr/yet-another-bench-script
For network speed using Ookla speedtest
https://network-speed.xyz/
https://bench.monster/
Thanks, this is exactly what I was looking for.
YABS is pretty good to measure disk and CPU performance (GB5 and GB6), but for network, i would go with Network-Speed.xyz, as it has regional speedtests.
and better to run it 2-3 times and on different time, maybe in the morning and evening, and take the average of them all, at least that's what i do while i benchmark the VPS.
"Piotr has done a LOT of benchmarking. How does he do it? The Phoronix test suite! This amazing free product offers HUNDREDS (thousands?) of benchmarks for different services, applications, and systems. In this video, Piotr walks through all the options and demos the world's most comprehensive benchmarking suite."
PTS is very good too, it has plenty tests and even can make our own scenarios. I used to use PTS as my benchmark methods, but it's often crashing or error in the middle of test.
The best benchmark is one designed for your actual use case. For example, measure throughput / requests per seconds to your web services.
Synthetic benchmarks are okay but nowhere near as good as real-world tests. One server can look faster than another when you use YABS, but the "slower" one might actually handle your workload better.