Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In Register

Quick Links


Categories

In this Discussion

› › Test the disk I/O of your VPS - Page 14
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.

Test the disk I/O of your VPS

1…8910111214»

Comments

  • qhosterqhoster Member

    This is totally idiotic running dd benchmarks and heavy test things on a productive node. You do not need and cannot expect 800MB.s for $5 VPS. And not sure why is needed with a drive space the VPS is offering 20GB or 30GB only ...

  • TarZZ92TarZZ92 Member

    one could also trick these I/O benchmarks too. very easily.

  • teknolaizteknolaiz Member
    edited August 2015

    IDE oldskool.

    # dd if=/dev/zero of=test_$$ bs=64k count=16k conv=fdatasync && rm -f test_$$
16384+0 Datensätze ein
16384+0 Datensätze aus
1073741824 Bytes (1,1 GB) kopiert, 345,861 s, 3,1 MB/s
  • MaouniqueMaounique Host Rep, Veteran

    @Hidden_Refuge said:
    IDE oldskool.

    > # dd if=/dev/zero of=test_$$ bs=64k count=16k conv=fdatasync && rm -f test_$$
> 16384+0 Datensätze ein
> 16384+0 Datensätze aus
> 1073741824 Bytes (1,1 GB) kopiert, 345,861 s, 3,1 MB/s
>

    Quantum bigfoot? :o

  • teknolaizteknolaiz Member

    Maounique said: Quantum bigfoot? :o

    A freebie I got. It's KVM VPS.
    Node: Dual E5-2650v3 with DDR4 RAM and RAID10 or so (don't remember exactly).
    IDE driver. I asked the provider to change to Virtio. Now on Virtio. Will post a result with Virtio later.

  • dragon1993dragon1993 Member

    @TarZZ92 said:
    one could also trick these I/O benchmarks too. very easily.

    60MB/s for 20$/mo (DO) VPS is terrible

  • k0nslk0nsl Member

    RamNode ;)

  • teknolaizteknolaiz Member

    Maounique said: Quantum bigfoot? :o

    Virtio:

    dd if=/dev/zero of=test_$$ bs=64k count=16k conv=fdatasync && rm -f test_$$
16384+0 Datensätze ein
16384+0 Datensätze aus
1073741824 Bytes (1,1 GB) kopiert, 5,58565 s, 192 MB/s
  • virzovirzo Member

    DB 1GB

    [root@server ~]# dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync

    16384+0 records in
    16384+0 records out
    1073741824 bytes (1.1 GB) copied, 5.85326 s, 183 MB/s

  • jazz1611jazz1611 Member
    edited September 2015

    ProfVPS, SSD KVM-02 (opening in PA, USA)

    [root@test ~]# dd if=/dev/zero of=test_$$ bs=64k count=16k conv=fdatasync && rm -f test_$$
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 1.23768 s, 868 MB/s
[root@test ~]# dd if=/dev/zero of=test_$$ bs=64k count=16k conv=fdatasync && rm -f test_$$
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 1.3207 s, 813 MB/s
  • inthecloudbloginthecloudblog Member

    VPS Dime is now working nicely:

    dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 5.56212 s, 193 MB/s
  • abangnoobabangnoob Member

    ramnode premium ovz 512mb:
    dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync16384+0 records in 16384+0 records out 1073741824 bytes (1.1 GB) copied, 1.08737 s, 987 MB/s

  • ClouviderClouvider Member, Patron Provider

    @inthecloudblog and the point of resurrecting the thread from 2011 was to... ?

    Not to mention that this (DD testing) is a pointless exercise, right ?

    Thanked by 1nfn
  • k0nslk0nsl Member

    @Clouvider isn't it better to contain this sort of "content" in one topic rather to have new posts for every single topic that features such content? The resurrection is justified in my view.

    Thanked by 2inthecloudblog HostSlick
  • inthecloudbloginthecloudblog Member

    I wasn't aware this post was not resurrected by someone else so I've posted. Also agree with @k0nsl

  • bersybersy Member

    I'm dare to ressurect it)

    Bandwagon $10/yr VPS (quite a bang for the buck)

    dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 2.71302 s, 396 MB/s
    Thanked by 2Tom yomero
  • david_Wdavid_W Member

    Thread from 2011?

  • bersybersy Member

    Yep!

  • noamannoaman Member 
    dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 1.20569 s, 891 MB/s

    Virmach 512 MB germany Frankfurt

  • HostingSpecialistsHostingSpecialists Member
    edited April 2016

    root@ukvps1:~# dd if=/dev/zero of=test bs=16k count=64k conv=fdatasync;rm -f test65536+0 records in
    65536+0 records out
    1073741824 bytes (1.1 GB) copied, 5.62438 s, 191 MB/s

  • mik997mik997 Member

    dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync 16384+0 records in 16384+0 records out 1073741824 bytes (1.1 GB) copied, 4.47889 s, 240 MB/s

    self-hosted vSphere6 VPS .. ;)

  • sinsin Member
    edited April 2016

    MikroVPS XEN - Hungary

    root@hu-server:~# dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync; rm test
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 2.85883 s, 376 MB/s
    Thanked by 1bersy
  • twaintwain Member

    Google Cloud Platform, CentOS7 KVM f1-micro instance, standard persistent disk, us-east1-b (South Carolina):

    Disk Speed
----------
I/O (1st run)   : 37.4 MB/s
I/O (2nd run)   : 37.6 MB/s
I/O (3rd run)   : 37.5 MB/s
Average I/O : 37.5 MB/s

$ dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync; rm test
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 28.6468 s, 37.5 MB/s

    Pretty good?? :) Also, I'm pretty sure the freevps bench script network speed tests used up my 1G free tier transfer allotment, lol.

  • jsgjsg Member, Resident Benchmarker

    Side note/hint: Those 'dd' tests de facto measure a lot of things many of which are not what you want. Plus running dd or similar for multiple seconds or worse disturbs the other VPSs on a node.

    A better alternative: vpsb -d -p /path_to_test

    That behaves nicely on a VPS and offers better results too e.g. sequential vs (true) random access for both reading and writing.

    Thanked by 1HostSlick
  • kerus1024kerus1024 Member

    something vps in lowendbox
    100GB HDD

    dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 4.89093 s, 220 MB/s

2nd
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 3.31921 s, 323 MB/s

3rd
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 3.75754 s, 286 MB/s

    and 100~170iops

    fast!

  • dragon1993dragon1993 Member

    ITLdc

    root@vps:~# dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync; rm test
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB, 1.0 GiB) copied, 6.71113 s, 160 MB/s

    root@vps:~# ioping -c 20 /
--- / (ext4 /dev/vda2) ioping statistics ---
20 requests completed in 19.0 s, 7.36 k iops, 28.8 MiB/s
min/avg/max/mdev = 102 us / 135 us / 187 us / 23 us
  • wavecomaswavecomas Member, Host Rep

    WaveCom OpenVZ

    [root@test ~]# dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync 16384+0 records in 16384+0 records out 1073741824 bytes (1.1 GB) copied, 1.05886 s, 1.0 GB/s [root@test ~]# [root@test ~]# dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync 16384+0 records in 16384+0 records out 1073741824 bytes (1.1 GB) copied, 1.12371 s, 956 MB/s [root@test ~]# dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync 16384+0 records in 16384+0 records out 1073741824 bytes (1.1 GB) copied, 1.15765 s, 928 MB/s [root@test ~]#

This discussion has been closed.