Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Speeds are much lower then expected #24

Open
ulfklose opened this issue Apr 18, 2023 · 1 comment
Open

Speeds are much lower then expected #24

ulfklose opened this issue Apr 18, 2023 · 1 comment

Comments

@ulfklose
Copy link

Hi, I'm using a Kingston NV2, which is a PCIe 4 x4 SSD, on a system with a PCIe3 x4 interface. The SSD is able to do 3.5 GB/s read and 2.8 GB/s write. This tool averages on about 500 MB/s which seems odd to me.

hdparm (sudo hdparm -tT --direct /dev/nvme0n1) results are about 1.7 GB/s (read), GNOME's disk utility shows about 2.3 GB/s.

Am I misinterpreting the results of your tool or aren't they comparable at all?

I'd love to see something similar to the AS SSD Benchmark for Windows on Linux.

@lucaspar
Copy link

lucaspar commented Jun 7, 2023

Numbers from different benchmarks can't be directly compared for several reasons, but the numbers I got from ssd-benchmark seem close to fio's when using a block size of 4k. Larger block sizes tend to yield larger speeds on benchmarks as well, perhaps as a result of reducing IOPS.

The closest way to replicate these Windows tools that report such high numbers was this script, also posted in this thread of someone trying to replicate CrystalDiskMark results on Linux.

That script gave me 5870MB/s sequential read on a Samsung 980 PRO PCIe 4.0 NVMe M2. I'd say that's in the same ballpark as the CrystalDiskMark screenshots I've seen, considering I was running the OS and many services on top as the benchmark executed.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants