[DiskSpeed, hdd/ssd benchmarking (unRAID 6+), version 2.10.8]

Hi, 

 

I really appreciate that you took the time to take a look. It's not a big deal; I imagine that it works as intended otherwise. 

 

Thanks again! 

[DiskSpeed, hdd/ssd benchmarking (unRAID 6+), version 2.10.8]

9 hours ago, jbartlett said:

Can you create a regular debug file and email it to me? Link and info is on the bottom of the DiskSpeed app screen. This way I can look at what the system is reporting for the drive.

Email has been sent with the regular debug file.

 

 

[DiskSpeed, hdd/ssd benchmarking (unRAID 6+), version 2.10.8]

3 hours ago, jbartlett said:

 

It's odd that the drive reported different speeds. The Samsung PM1733 has a max read speed of 6,400 MB/s based on this spec video but your speeds are much slower for the SSD. Only recommendation I have is to verify that you have it connected to a PCIe Gen 4 interface. It kinda looks like a gen 3 speed.

I found that odd as well that it had two different speeds.

 

To shed more light, I am using an ASRock RomeD8-2T motherboard, it has 7 PCie 4.0 slots. I have the Samsung PM1733 ssd connected to slot 1 using a U.2 to PCie adapter (https://www.amazon.com/gp/product/B01D2PXUAQ/ref=ppx_yo_dt_b_asin_title_o02_s00?ie=UTF8&psc=1) . It states it should support PCie 4.0 speeds if the slot supports it; and as previously mentioned, the windows VM reported the expected speeds so I wouldn't think the interface would be the limit.

I am conflicted with the results from the different tests ranging from 800MB/s in the disk speed docker, all the way to 6GB/s+ within a windows VM which makes it difficult to know which result to trust and use as a baseline to troubleshoot. I could potentially try another adapter if you have any solid recommendations.

[DiskSpeed, hdd/ssd benchmarking (unRAID 6+), version 2.10.8]

11 hours ago, jbartlett said:

Apologies, I misread your question and my brain flipped the numbers.

 

If you don't mind, let's try two tests against the drive. The first test duplicates how DiskSpeed performs a benchmark at the start. In these examples, I'm testing against /dev/sdd - change the drive reference to reflect your PM1733. This first test you will need to break out to stop.

Test 1: dd if=/dev/sdd of=/dev/null bs=1310720 skip=0 iflag=direct conv=noerror status=progress

 

Now let's do the same but against a file. Locate a large file and change the if= reference to point to it.

Test 2: dd if=/mnt/disks/SSD_327E/Cam1/OS.qcow2 of=/dev/null bs=1310720 skip=0 iflag=direct conv=noerror status=progress

 

Test 2 is how I plan to have version 3 of DiskSpeed to perform a benchmark. What are the transfer speeds?

Ah, thank you for following up then!
 

I am not sure how long I needed to run the test so I stopped when there was no improvement.

Here are the results of test 1:
root@Tower:~# dd if=/dev/nvme0n1p1 of=/dev/null bs=1310720 skip=0 iflag=direct conv=noerror status=progress
43972034560 bytes (44 GB, 41 GiB) copied, 32 s, 1.4 GB/s^C
33920+0 records in
33919+0 records out
44458311680 bytes (44 GB, 41 GiB) copied, 32.3474 s, 1.4 GB/s

 

Here are the results of test 2:

root@Tower:~# dd if=/mnt/disks/Sam4tb/Test.mkv of=/dev/null bs=1310720 skip=0 iflag=direct conv=noerror status=progress
22972989440 bytes (23 GB, 21 GiB) copied, 11 s, 2.1 GB/s
18570+1 records in
18570+1 records out
24340223273 bytes (24 GB, 23 GiB) copied, 11.5635 s, 2.1 GB/s
 

Here is the drive listed using 'df'
/dev/nvme0n1p1 3748905808   49940604 3698965204   2% /mnt/disks/Sam4tb

 

I hope I ran these tests correctly. Please let me know if these results are expected or if there is anything else I can provide,

 

 

[DiskSpeed, hdd/ssd benchmarking (unRAID 6+), version 2.10.8]

Thank you for your feedback

 

The pm1733 drive had been formatted prior to the test if that makes a difference. I'll just assume that the drive works as intended for now. 

[DiskSpeed, hdd/ssd benchmarking (unRAID 6+), version 2.10.8]

Pardon if this is a silly question, but how reliable are the figures provided by the benchmark?

 

I have two nvme SSDs and one has results which match its specs, while the other is horrible.

 

Samsung 970 Evo Plus shows between 2500-3000 MB/s which is expected

 

Samsung PM1733 shows between 700-850MB/s which is horrible compared to its claimed 7000MB/s spec. The odd thing is that I passed it through to a windows VM and ran AS ssd benchmark and showed reasonable results around 6700MB/s.

 

I am just curious to know whether its an error of the plugin and just ignore its result or not.

Thank you in advance