Use of Diskspd for stress testing about diskspd HOT 6 CLOSED

microsoft commented on May 23, 2024

Use of Diskspd for stress testing

from diskspd.

Comments (6)

dl2n commented on May 23, 2024

“Behaving well” is not a well defined test. All DISKSPD does is what it says it is doing; in this case * 256KB IO * 1 thread * 100% write * 5 second warmup * up to 64 IOs outstanding * unbuffered writethrough IO * random IO pattern across the target at IO size alignment (256KB) * run for 600 seconds * 20MB random write source buffer (each write buffer will be random chosen from this, at 4 byte random alignment) * Gather latency data * An apparent syntax error which DISKSPD may ignore re: 0_pct_comp.bin (you may have intended this to be put against -Z20M, in which case the write source buffer would have been initialized from that file) * run v. PhysicalDisk2 If this is what you wanted, you have to use the statistics to reason about goodness. I don’t know what you consider good. In general, 193MB/s indicates the drive is reordering the queued IO into a good elevator sweep across the HDD. The 64QD attempt is responsible for that, at a cost of about 84ms per IO and a large standard deviation of 42ms. The latency is actually lower than I would have expected all things considered. The latency percentiles seem pretty much in control given the worst case is only 170ms or so. The real question to ask is whether this is the right way to characterize the device for the workload you want to apply to it. Sent from Mail<https://go.microsoft.com/fwlink/?LinkId=550986> for Windows 10

…

________________________________ From: v-sankod <[email protected]> Sent: Thursday, March 15, 2018 11:50:48 AM To: Microsoft/diskspd Cc: Subscribed Subject: [Microsoft/diskspd] Use of Diskspd for stress testing (#56) Hi all, I am not sure if this is the right platform for me to address my requirement. I am looking for just stressing out the HDD/SSD by doing 100% write to the disk and checking some basic info as to how the disk is enduring during stress. I am are not looking for performance measurement here. Just want to run diskspd for ~30 minutes and check if the disk is behaving well during stress. Please let me know what test I should perform and what result I should be looking for ? Kindly help me :) I tried writing to the disk (100% write for 10 mins ) using the following command diskspd.exe -b256K -t1 -w100 -W5 -o64 -Swu -r -d600 -Z20M -L,0_pct_comp.bin #2<#2> And got the output like so : Total IO thread | bytes | I/Os | MB/s | I/O per s | AvgLat | LatStdDev | file 0 | 121647398912 | 464048 | 193.35 | 773.40 | 82.740 | 41.172 | #2 (894GB)

________________________________ total: 121647398912 | 464048 | 193.35 | 773.40 | 82.740 | 41.172 Read IO thread | bytes | I/Os | MB/s | I/O per s | AvgLat | LatStdDev | file 0 | 0 | 0 | 0.00 | 0.00 | 0.000 | N/A | #2 (894GB)

________________________________ total: 0 | 0 | 0.00 | 0.00 | 0.000 | N/A Write IO thread | bytes | I/Os | MB/s | I/O per s | AvgLat | LatStdDev | file 0 | 121647398912 | 464048 | 193.35 | 773.40 | 82.740 | 41.172 | #2 (894GB)

________________________________ total: 121647398912 | 464048 | 193.35 | 773.40 | 82.740 | 41.172 %-ile | Read (ms) | Write (ms) | Total (ms) min | N/A | 31.242 | 31.242 25th | N/A | 33.822 | 33.822 50th | N/A | 96.928 | 96.928 75th | N/A | 103.014 | 103.014 90th | N/A | 153.568 | 153.568 95th | N/A | 156.915 | 156.915 99th | N/A | 160.514 | 160.514 3-nines | N/A | 163.477 | 163.477 4-nines | N/A | 165.460 | 165.460 5-nines | N/A | 167.004 | 167.004 6-nines | N/A | 168.096 | 168.096 7-nines | N/A | 168.096 | 168.096 8-nines | N/A | 168.096 | 168.096 9-nines | N/A | 168.096 | 168.096 max | N/A | 168.096 | 168.096 — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub<#56>, or mute the thread<https://github.com/notifications/unsubscribe-auth/AIVpKV6Nbrw6ZNvb12e6cNRx-8pyJbc4ks5tergHgaJpZM4Sssub>.

from diskspd.

v-sankod commented on May 23, 2024

Thanks for your reply.
From the reply, I can understand that after running Diskspd, we can only expect performance values which can be used for benchmarking/comparing and analyzing the statistics.
In our project, we are just looking to stress a disk by running Diskspd and then we will be collecting the SMART attributes to find any SMART faults.

I have one more question:
Can we not run diskspd on disks which is formatted as NTFS/ReFS/FAT32 ?
When I try that, I get the error : error during overlapped IO operation (error code: 5)

This is the Diskspd command I run:
DiskSpd.exe -b256K -t1 -w100 -W15 -o64 -Swu -r -d10 -v -Z20M,0_pct_comp.bin #5

from diskspd.

dl2n commented on May 23, 2024

You are probably not administrator (5 == ERROR_ACCESS_DENIED). However, if you run a write test on the raw physical device (#5), it will do so – and irretrievably corrupt whatever is on the device, including the filesystem. This is how chkdsk works :-) If you want to run through the filesystem, mount it and create a file to run DISKSPD against. Sent from Mail<https://go.microsoft.com/fwlink/?LinkId=550986> for Windows 10

…

________________________________ From: v-sankod <[email protected]> Sent: Wednesday, March 21, 2018 6:22:42 PM To: Microsoft/diskspd Cc: Dan Lovinger; Comment Subject: Re: [Microsoft/diskspd] Use of Diskspd for stress testing (#56) Thanks for your reply. From the reply, I can understand that after running Diskspd, we can only expect performance values which can be used for benchmarking/comparing and analyzing the statistics. In our project, we are just looking to stress a disk by running Diskspd and then we will be collecting the SMART attributes to find any SMART faults. I have one more question: Can we not run diskspd on disks which is formatted as NTFS/ReFS/FAT32 ? When I try that, I get the error : error during overlapped IO operation (error code: 5) This is the Diskspd command I run: DiskSpd.exe -b256K -t1 -w100 -W15 -o64 -Swu -r -d10 -v -Z20M,0_pct_comp.bin #5<#5> — You are receiving this because you commented. Reply to this email directly, view it on GitHub<#56 (comment)>, or mute the thread<https://github.com/notifications/unsubscribe-auth/AIVpKVn322Nfj92LtMFWTYbGNKpXn9I6ks5tgvzigaJpZM4Sssub>.

from diskspd.

v-sankod commented on May 23, 2024

Thanks for your quick response.
I had ran diskspd in admin mode.
These are the steps I followed:

diskpart> select disk 1
diskpart> create partition primary
diskpart> format fs=ntfs quick

Ran DiskSpd.exe -b256K -t1 -w100 -W15 -o64 -Swu -r -d10 -Z20M,0_pct_comp.bin #1
error during overlapped IO operation (error code: 5)
There has been an error during threads execution

Command Line: DiskSpd.exe -b256K -t1 -w100 -W15 -o64 -Swu -r -d10 -Z20M,0_pct_comp.bin #1

Input parameters:

    timespan:   1
    -------------
    duration: 10s
    warm up time: 15s
    cool down time: 0s
    random seed: 0
    path: '#1'
            think time: 0ms
            burst size: 0
            software cache disabled
            hardware write cache disabled, writethrough on
            write buffer size: 20971520
            write buffer source: '0_pct_comp.bin'
            performing write test
            block size: 262144
            using random I/O (alignment: 262144)
            number of outstanding I/O operations: 64
            thread stride size: 0
            threads per file: 1
            using I/O Completion Ports
            IO priority: normal

Results for timespan 1:

The test was interrupted before the measurements began. No results are displayed.
Error generating I/O requests

from diskspd.

dl2n commented on May 23, 2024

What was the error that time?

from diskspd.

dl2n commented on May 23, 2024

No response, closing.

from diskspd.

Use of Diskspd for stress testing about diskspd HOT 6 CLOSED

Comments (6)

Related Issues (20)

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent