Jump to content

KDiskMark -- real world test run results.


abarbarian

Recommended Posts

Benchmark Storage Drives in Linux with KDiskMark

 

Quote

KDiskMark is a free and open source graphical benchmarking utility that can be used to measure performance of storage drives. It can determine read and write speed estimates with good accuracy for both hard drives and solid state drives. It can run benchmarks on external drives connected to your Linux system as well.

 

The above article is a good read for folk new to this program.

 

https://github.com/JonMagon/KDiskMark

 

Quote

KDiskMark is an HDD and SSD benchmark tool with a very friendly graphical user interface. KDiskMark with its presets and powerful GUI calls Flexible I/O Tester and handles the output to provide an easy to view and interpret comprehensive benchmark result. The application is written in C++ with Qt and doesn't have any KDE dependencies.

 

Nice to see that there is a Qt program with no KDE dependencies.

 

When I first ran the program it only showed me my nvme drives and I thought it only dealt with that type. When sunrat posted that he had run the program on his ssd's it made me think how. The answer is quite simple. Duh your drives have to be mounted for KDiskMark to see them.

Some of the articles regarding KDiskMark say that the "PROFILE" section has not been implemented fully yet. I can report that this section is working now.

 

A point to note is that running tests on your working os drive will not give fully true results as to your drives performance. The best way to find out true results for your os drive is to run KDiskMark from a Live OS.

 

So that we can get some real information here when testing and posting a screenshot could folk post results from tests run with the DEFAULT setting in the PROFILE section.  If you use a different setting post that information along with the screenshot. From the screenshot we can see whether you have done a 1GB or 32GB etc run and we can also see how many runs the test had.

 

4wCSKwQ.png

 

I also note that there is a STANDARD PRESET and a NVME SSD setting which I have no information on at the moment. Does that setting mean you test nvme drives with e NVME SSD setting and other ssd's on the STANDARD PRESET ? More information needed here.

 

K93jmsJ.png

 

Just for a bit of fun here are some quick tests I did on a 500GB ssd that only had 24GB of free space. A 250GB ssd that is empty. An a half full 250GB nvme running in a usb external enclosure.

 

Almost full 500GB

 

6oNYk8M.png

 

Empty 250GB

 

JPqYkSa.png

 

Empty 250GB ten minutes later. Note the 50% jump in the result at bottom left.

 

zz2GSiT.png

 

Interestingly the two drives post almost matching results apart from the anomaly.

 

NVME 250GB from usb

 

myqQaJ0.png

 

Interestingly the top two results are a lot lower than the ssd's connected to the motherboard with the RND4k (IOPS) way way higher.This is a most puzzling result. All the results are way way lower than the test I did on the nvme's connected to the motherboard,apart from the RND4K results on the bottom row which were almost three times higher.

 

rAJ7mni.png

 

Have fun folks. 😋

 

The STANDARD PRESET and a NVME SSD setting do make a difference.

 

STANDARD PRESET

 

XsZVvqv.png

 

NVME SSD setting

 

n4iEz8Q.png

 

Though the results are rather strange. 🤔

Edited by abarbarian
  • Like 1
Link to comment
Share on other sites

No question that if your motherboard has the capability, NVME is the way to go - especially for larger sequential files, such as what you'd get in booting, gaming, video rendering.

My Dell desktop has only support for 1 NVME drive so I've got that set up for my train sims and O/S. I have a large HDD for data and two prewired bays for additional SATA SSDs if I want.

The rest of my hardware is strictly SATA based but most of the time that's OK. I am usually fixing up old laptops which had mechanical drives so a SATA SSD swap is usually easy enough and does make a difference. SATA SSDs are really cheap here in Canada as well - you can get a 1 TB for $50 Canadian - about 30 GBP.

  • +1 1
Link to comment
Share on other sites

4 minutes ago, raymac46 said:

No question that if your motherboard has the capability, NVME is the way to go - especially for larger sequential files, such as what you'd get in booting, gaming, video rendering.

My Dell desktop has only support for 1 NVME drive so I've got that set up for my train sims and O/S. I have a large HDD for data and two prewired bays for additional SATA SSDs if I want.

The rest of my hardware is strictly SATA based but most of the time that's OK. I am usually fixing up old laptops which had mechanical drives so a SATA SSD swap is usually easy enough and does make a difference. SATA SSDs are really cheap here in Canada as well - you can get a 1 TB for $50 Canadian - about 30 GBP.

 

When you get the time why not post a couple of test results. :hug:

Link to comment
Share on other sites

Just for a laugh I decided to test my HDD a 500GB Samsung  HD502HJ that is three-quarters full and running from a USB2 dock. I set it running and went to have lunch and had time to have a shower before it had finished the run.

 

EUnTrjg.png

 

🤣

 

To be fair when backing up to the drive with rsync it does not seem like watching paint dry. 😎

Link to comment
Share on other sites

Well here is the result from my Windows boot drive on an ASUS laptop. Obviously results are skewed because the drive is in use. Looks like it is NVME tho.

 

 

CrystalDiskMark_20230611093058.png

  • Like 1
Link to comment
Share on other sites

This is a cheap $50 SSD in an old Broadwell laptop. Nothing to brag about, but it made quite a difference on how the old lappy runs.

 

Screenshot_2023-06-11_10-15-59.png

Edited by raymac46
  • Like 1
Link to comment
Share on other sites

Ugly old Toshiba SSD on a Linux Mint desktop powered by an 11 year old AMD Piledriver CPU. Still runs the system like a champion for most uses.

 

 

Screenshot from 2023-06-11 10-40-09.png

Link to comment
Share on other sites

V.T. Eric Layton

There's a similar, but simpler, app like this in Porteus 2.0 disk that I have. It's called "Disk Check", I think. Works very well. Gives info on the important specs and performance of my drives. I use it occasionally to check "disk health".

 

For some reason or another, they did not include in in Porteus 3, which is why I still have the v2 disk.

Link to comment
Share on other sites

I'm guessing with good SATA drives, they will always read 500-550MB/s as that is limited by the bus bandwidth.

 

WD Blue 1TB SATA m.2 SSD - my main system drive

KDM-WDBlue1TBSSD.png

 

Samsung EVO850 500GB SATA SSD - data

KDM-Samsung500GBEVO850.png

 

WD Blue 4TB SATA HD - data drive

KDM-WDBlue4TBHD.png

Link to comment
Share on other sites

On 6/11/2023 at 6:11 PM, V.T. Eric Layton said:

For some reason or another, they did not include in in Porteus 3, which is why I still have the v2 disk.

 

Yeah I found out that it is no longer included in Porteus. You can install it on the Live os and of course you can keep it installed if you have persistence enabled. Strange that it was dropped though as it is only 1.4 MB so it can not be to save space on the os.🤔

  • Thanks 1
Link to comment
Share on other sites

On 6/11/2023 at 3:42 PM, raymac46 said:

Ugly old Toshiba SSD on a Linux Mint desktop powered by an 11 year old AMD Piledriver CPU. Still runs the system like a champion for most uses.

 

 

Screenshot from 2023-06-11 10-40-09.png

 

Those are all pretty reasonable results considering the hardware. 🤩

Link to comment
Share on other sites

On 6/12/2023 at 12:55 AM, sunrat said:

I'm guessing with good SATA drives, they will always read 500-550MB/s as that is limited by the bus bandwidth.

 

 

WD Blue 4TB SATA HD - data drive

KDM-WDBlue4TBHD.png

 

Decent results apart from the dismal showing for the RND4K stuff on the 4 TB 🙀

On 6/12/2023 at 12:55 AM, sunrat said:

I'm guessing with good SATA drives, they will always read 500-550MB/s as that is limited by the bus bandwidth.

 

Recon you are spot on there. 😎

Link to comment
Share on other sites

0YdwBXU.png

 

Tough Book CF52 Intel Core2 Duo T7300 with a dirt cheap four year old 120 GB ssd.

 

As you can see I have installed my favourite Window Maker. 😃

 

Link to comment
Share on other sites

3 hours ago, abarbarian said:

Decent results apart from the dismal showing for the RND4K stuff on the 4 TB 🙀

 Yeah, seek times for HDDs are in a different universe from SSDs.

 

Your SATA USB2 dock would definitely be holding back the speed of your HDD posted above. I have several of those same WD Blue 4TB drives I posted in USB3 external cases and the speed is close to the internal SATA3 drive.

Link to comment
Share on other sites

8 hours ago, sunrat said:

Yeah, seek times for HDDs are in a different universe from SSDs.

 

Ah ha I missed that the 4TB was a HD.

 

The dock is a twin slot and I only use it for storage. If I need to access a file it does not take that long to transfer to my main pc. Anything else I need to do with it I just set it running and go and have a cuppa .An possibly a scone with jam and clotted cream. 😋

Link to comment
Share on other sites

Your Toughbook likely has SATA2 capability which will impact the SSD performance. Still better than a mechanical drive though.

  • Agree 1
Link to comment
Share on other sites

This is an interesting article on nvmes on RAID and with LUKS,

 

PCIe Bifurcation and NVMe RAID in Linux Part 1: The Hardware

 

kdiskmark_sn770_encryption.png

 

kdiskmark_raid1-1.png

 

Quote

Notice that the type of KDiskMark preset used affects the numbers. NVMe preset yields higher numbers, but the Q1T1 number from the standard preset is what we want to watch the most since it tends to best reflect real-world performance during everyday usage.

 

nvme_single_raid.png

 

I know that this next article is a bit of a diversion from the main part of this thread 😜

 

Another very interesting article concerns speed tests on various usb sticks used to run LIVE OS'S. They were testing Mageia but I can not see why the results would be much different for other os's.

 

https://wiki.mageia.org/en/Storage_speed_test_results

 

Quote
  • Of the media I have tested most was already used, but even the new ones did not get notably affected in write performance after being fed their capacity of data, forcing erase. Probably all(?) have lots of spare space??
  • Two of the USB stick was not accepted as bootable media on one of my Laptops!
  • High end USB sticks beats a mainstream(?) external SSD in speed by good margin!

 

😎

Link to comment
Share on other sites

Interesting article including USB stick data. I just tested a Sandisk Extreme 64GB. It writes a Debian netinstall image of 738MB in 9 seconds.

 

KDiskMark default test:

sandisk-extreme-ext4.png.996977b9afbbae7e8ca4078dbbf7d77c.png

Link to comment
Share on other sites

3 hours ago, sunrat said:

Interesting article including USB stick data. I just tested a Sandisk Extreme 64GB. It writes a Debian netinstall image of 738MB in 9 seconds.

 

Yes I found it interesting it surprised  me how some big name drives were so slow.

 

Your  Sandisk is a USB3 I guess as it is miles quicker than my USB2 Dock. 9 secs is pretty nippy. 😎

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...