Howdy, Stranger!

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


KS-LE1 port speed
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.

KS-LE1 port speed

ralfralf Member

Don't want to necro the old thread, so created a new one for a fairly trivial comment.

When I got my KS-LE1 over black friday, I YABS'd it at the time and found I hadn't been lucky enough to get one with the faster network port as all of my speeds were around 95-97 Mbps.

Last night I happened to YABS one of the VMs I created on it because I wanted to check the disk IO speed and noticed that my IP4 receive speeds were all over 500Mbps to the European networks.

Anyone else noticed anything like this before?

Comments

  • Just use it, don't announce :D

  • @jason5545 said:
    Just use it, don't announce :D

    This :)

  • jmgcaguiclajmgcaguicla Member
    edited April 2022

    LET has been infiltrated by KS glowies.

    Do not be tricked into snitching on your port speed.

  • @ralf Please delete this, they are here spying ...

  • VoidVoid Member

    Dear customer, thanks for letting us know. It was a glitch and your speed has been reduced to 95Mbps.

    Thanked by 1icarus11
  • henixhenix Member
    iperf3 Network Speed Tests (IPv4):
    ---------------------------------
    Provider        | Location (Link)           | Send Speed      | Recv Speed
                    |                           |                 |
    Clouvider       | London, UK (10G)          | 95.2 Mbits/sec  | 874 Mbits/sec
    Online.net      | Paris, FR (10G)           | 95.4 Mbits/sec  | 894 Mbits/sec
    WorldStream     | The Netherlands (10G)     | busy            | 858 Mbits/sec
    WebHorizon      | Singapore (400M)          | busy            | busy
    Clouvider       | NYC, NY, US (10G)         | 91.0 Mbits/sec  | 347 Mbits/sec
    Velocity Online | Tallahassee, FL, US (10G) | 87.2 Mbits/sec  | 344 Mbits/sec
    Clouvider       | Los Angeles, CA, US (10G) | 87.2 Mbits/sec  | 218 Mbits/sec
    Iveloz Telecom  | Sao Paulo, BR (2G)        | busy            | busy
    
    iperf3 Network Speed Tests (IPv6):
    ---------------------------------
    Provider        | Location (Link)           | Send Speed      | Recv Speed
                    |                           |                 |
    Clouvider       | London, UK (10G)          | 94.0 Mbits/sec  | 846 Mbits/sec
    Online.net      | Paris, FR (10G)           | 93.9 Mbits/sec  | 850 Mbits/sec
    WorldStream     | The Netherlands (10G)     | busy            | busy
    WebHorizon      | Singapore (400M)          | busy            | busy
    Clouvider       | NYC, NY, US (10G)         | 89.8 Mbits/sec  | 313 Mbits/sec
    
    Thanked by 1ralf
  • ralfralf Member

    I was fine with 100 Mbps on this server anyway as it doesn't see all that much traffic, it was just more of a surprise that the port speed seemingly increased.

  • @ralf said:
    I was fine with 100 Mbps on this server anyway as it doesn't see all that much traffic, it was just more of a surprise that the port speed seemingly increased.

    100 Mbps is ok for general use indeed.

  • ErisaErisa Member

    The network speed is determined and throttled by their routers.

    Sometimes their routers fuck up. Some servers might have no speed increase, some might have speed increase over IPv6, and some servers might have speed increase against certain other remote targets (It's supposed to have internal traffic be faster but doesn't always and sometimes other same region traffic gets caught up)

    All of these are mistakes and not intentional. When OVH realises, they revert them.
    I haven't heard of them making that mistake fresh after not doing it before, but in all honesty I'm not surprised.

    Enjoy your speed increase. Just know. it will not be there forever.

    Thanked by 2jason5545 ralf
  • @Erisa said:
    The network speed is determined and throttled by their routers.

    Sometimes their routers fuck up. Some servers might have no speed increase, some might have speed increase over IPv6, and some servers might have speed increase against certain other remote targets (It's supposed to have internal traffic be faster but doesn't always and sometimes other same region traffic gets caught up)

    All of these are mistakes and not intentional. When OVH realises, they revert them.
    I haven't heard of them making that mistake fresh after not doing it before, but in all honesty I'm not surprised.

    Enjoy your speed increase. Just know. it will not be there forever.

    Let me think that a guy on LET is trying to sell a Kimsufi server with dual 1G ports, he kinda treats it like a selling point, but it should be treated as a hidden benefit IMHO, because you never know when they will revert it back.

  • ErisaErisa Member
    edited April 2022

    @jason5545 said:

    @Erisa said:
    The network speed is determined and throttled by their routers.

    Sometimes their routers fuck up. Some servers might have no speed increase, some might have speed increase over IPv6, and some servers might have speed increase against certain other remote targets (It's supposed to have internal traffic be faster but doesn't always and sometimes other same region traffic gets caught up)

    All of these are mistakes and not intentional. When OVH realises, they revert them.
    I haven't heard of them making that mistake fresh after not doing it before, but in all honesty I'm not surprised.

    Enjoy your speed increase. Just know. it will not be there forever.

    Let me think that a guy on LET is trying to sell a Kimsufi server with dual 1G ports, he kinda treats it like a selling point, but it should be treated as a hidden benefit IMHO, because you never know when they will revert it back.

    Correct. Because it can be reverted at any time and that buyer might get mad.

    If I sold a Kimsufi server with increased bandwidth I would not factor that into the price and would merely mention it along with a disclaimer that it might go away.

    Thanked by 1jason5545
  • rm_rm_ IPv6 Advocate, Veteran
    edited April 2022

    @ralf said: When I got my KS-LE1 over black friday, I YABS'd it at the time and found I hadn't been lucky enough to get one with the faster network port as all of my speeds were around 95-97 Mbps.

    Are you sure you are not mis-remembering it? Maybe back then you did not YABS, but only the upload test? It is super common for these servers to have 100 Mbit upload and 1 Gbit download. Maybe even all of them do. But only a few have 1 Gbit upload as well (or on IPv6). And it's not as common for these limits to change over time after you get it.

  • ralfralf Member

    @rm_ said:
    Are you sure you are not mis-remembering it?

    That's totally possible, to be honest, but I was pretty sure it was 100Mbps each way.

  • @ralf said:

    @rm_ said:
    Are you sure you are not mis-remembering it?

    That's totally possible, to be honest, but I was pretty sure it was 100Mbps each way.

    Please share YABS

  • This seems to happen quite a lot, people receiving more HDD or bandwidth. Probably very uncommon though.

  • ralfralf Member
    edited April 2022

    @Sanjue007 said:

    @ralf said:

    @rm_ said:
    Are you sure you are not mis-remembering it?

    That's totally possible, to be honest, but I was pretty sure it was 100Mbps each way.

    Please share YABS

    This is a freshly created 2-core KVM instance, with 20GB striped FS. Swap is a 2GB file on instance's root filesystem. (Basically the same as I'm using for most of my instances, except they're all 10GB). Also no IPv6 as the instances either talk to each other or to other servers over wireguard, so I only enabled IPv4.

    ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ##

    Yet-Another-Bench-Script

    v2022-02-18

    https://github.com/masonr/yet-another-bench-script

    ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ##

    Fri Apr 29 15:29:37 UTC 2022

    Basic System Information:

    Processor : Intel Xeon E3-12xx v2 (Ivy Bridge, IBRS)
    CPU cores : 2 @ 3699.998 MHz
    AES-NI : ✔ Enabled
    VM-x/AMD-V : ❌ Disabled
    RAM : 974.5 MiB
    Swap : 2.0 GiB
    Disk : 19.9 GiB

    fio Disk Speed Tests (Mixed R/W 50/50):

    Block Size 4k (IOPS) 64k (IOPS)
    Read 1.03 MB/s (257) 14.45 MB/s (225)
    Write 1.06 MB/s (265) 15.02 MB/s (234)
    Total 2.09 MB/s (522) 29.47 MB/s (459)
    Block Size 512k (IOPS) 1m (IOPS)
    ------ --- ---- ---- ----
    Read 57.27 MB/s (111) 57.55 MB/s (56)
    Write 60.07 MB/s (117) 62.15 MB/s (60)
    Total 117.35 MB/s (228) 119.70 MB/s (116)

    iperf3 Network Speed Tests (IPv4):

    Provider | Location (Link) | Send Speed | Recv Speed
    | | |
    Clouvider | London, UK (10G) | 96.9 Mbits/sec | 722 Mbits/sec
    Online.net | Paris, FR (10G) | busy | busy
    WorldStream | The Netherlands (10G) | 96.8 Mbits/sec | 732 Mbits/sec
    WebHorizon | Singapore (400M) | busy | busy
    Clouvider | NYC, NY, US (10G) | 61.7 Mbits/sec | 101 Mbits/sec
    Velocity Online | Tallahassee, FL, US (10G) | 91.7 Mbits/sec | 72.6 Mbits/sec
    Clouvider | Los Angeles, CA, US (10G) | 91.0 Mbits/sec | 67.9 Mbits/sec
    Iveloz Telecom | Sao Paulo, BR (2G) | busy | busy

    Geekbench 5 Benchmark Test:

    Test | Value
    |
    Single Core | 863
    Multi Core | 1663
    Full Test | https://browser.geekbench.com/v5/cpu/14620907

  • KS-LEs are mostly 1000 down, 100 up... making them pretty useless for a lot of things.

  • @JoeMerit said:
    KS-LEs are mostly 1000 down, 100 up... making them pretty useless for a lot of things.

    2 in the same DC can do 1000 each way to each other.

  • ya but if you have a 1gbit connection at home, enjoy only getting 10% of your speed from your dedi.

Sign In or Register to comment.