Howdy, Stranger!

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


Strange! Why is Hetzner cloud much more expensive than Hetzner dedi? - Page 2
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.

Strange! Why is Hetzner cloud much more expensive than Hetzner dedi?

2»

Comments

  • nullroutenullroute Member, Host Rep

    40G port;
    Raid 10 disk redundancy;
    Ease of backups (automatic snapshots);
    USA location;
    Perfect uptime (No downtime on my US VM since its launch of the US location on Hetzner);
    It can easily be scalable;
    The nvme disks you get are top of the line (5x more performance than the disks offered on dedicated ones);
    And there is basic DDoS protection on cloud servers - there is NO DDOS PROTECTION on bare metal hetzner servers.

    Thanked by 1PulsedMedia
  • Clearly nobody here has actually benchmarked both, but the only valid claim is port speed (of course you can get 10gbps on dedis).

    But the CPU and disks are much slower on cloud, it’s not even comparable.

  • @nullroute said: 40G port;

    Not 10 Gbit/s-Port?

    @iwaswrongonce said: But the CPU and disks are much slower on cloud, it’s not even comparable.

    Compared to AX41?

  • PulsedMediaPulsedMedia Member, Patron Provider

    @nullroute said:
    40G port;
    Raid 10 disk redundancy;
    Ease of backups (automatic snapshots);
    USA location;
    Perfect uptime (No downtime on my US VM since its launch of the US location on Hetzner);
    It can easily be scalable;
    The nvme disks you get are top of the line (5x more performance than the disks offered on dedicated ones);
    And there is basic DDoS protection on cloud servers - there is NO DDOS PROTECTION on bare metal hetzner servers.

    good points, i didn't check specs that thoroughly.
    Doesn't look bad deal in that light at all

    @iwaswrongonce said:
    Clearly nobody here has actually benchmarked both, but the only valid claim is port speed (of course you can get 10gbps on dedis).

    But the CPU and disks are much slower on cloud, it’s not even comparable.

    we got a AX41, i could do that

  • PulsedMediaPulsedMedia Member, Patron Provider

    Yabs results:

    AX51-NVME (Raid1)

    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    #              Yet-Another-Bench-Script              #
    #                     v2023-03-24                    #
    # https://github.com/masonr/yet-another-bench-script #
    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    
    Fri 31 Mar 2023 03:34:38 PM CEST
    
    Basic System Information:
    ---------------------------------
    Uptime     : 186 days, 5 hours, 41 minutes
    Processor  : AMD Ryzen 7 3700X 8-Core Processor
    CPU cores  : 16 @ 2139.387 MHz
    AES-NI     : ✔ Enabled
    VM-x/AMD-V : ✔ Enabled
    RAM        : 62.8 GiB
    Swap       : 32.0 GiB
    Disk       : 906.2 GiB
    Distro     : Debian GNU/Linux 11 (bullseye)
    Kernel     : 5.10.0-18-amd64
    VM Type    : NONE
    Net Online : IPv4
    
    IPv4 Network Information:
    ---------------------------------
    ISP        : Hetzner Online GmbH
    ASN        : AS24940 Hetzner Online GmbH
    Host       : Hetzner
    Location   : Helsinki, Uusimaa (18)
    Country    : Finland
    
    fio Disk Speed Tests (Mixed R/W 50/50):
    ---------------------------------
    Block Size | 4k            (IOPS) | 64k           (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 625.93 MB/s (156.4k) | 1.21 GB/s    (18.9k)
    Write      | 627.58 MB/s (156.8k) | 1.21 GB/s    (19.0k)
    Total      | 1.25 GB/s   (313.3k) | 2.42 GB/s    (37.9k)
               |                      |                     
    Block Size | 512k          (IOPS) | 1m            (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 971.87 MB/s   (1.8k) | 1.01 GB/s      (991)
    Write      | 1.02 GB/s     (1.9k) | 1.08 GB/s     (1.0k)
    Total      | 1.99 GB/s     (3.8k) | 2.09 GB/s     (2.0k)
    
    iperf3 Network Speed Tests (IPv4):
    ---------------------------------
    Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping           
    -----           | -----                     | ----            | ----            | ----           
    Clouvider       | London, UK (10G)          | 934 Mbits/sec   | 259 Mbits/sec   | 36.3 ms        
    Scaleway        | Paris, FR (10G)           | busy            | 104 Mbits/sec   | 36.4 ms        
    NovoServe       | North Holland, NL (40G)   | 937 Mbits/sec   | busy            | 30.2 ms        
    Uztelecom       | Tashkent, UZ (10G)        | 915 Mbits/sec   | 119 Mbits/sec   | 60.8 ms        
    Clouvider       | NYC, NY, US (10G)         | 791 Mbits/sec   | 98.4 Mbits/sec  | 97.4 ms        
    Clouvider       | Dallas, TX, US (10G)      | 536 Mbits/sec   | 44.9 Mbits/sec  | 139 ms         
    Clouvider       | Los Angeles, CA, US (10G) | 175 Mbits/sec   | 21.3 Mbits/sec  | 170 ms         
    
    Geekbench 6 Benchmark Test:
    ---------------------------------
    Test            | Value                         
                    |                               
    Single Core     | 1717                          
    Multi Core      | 8641                          
    Full Test       | https://browser.geekbench.com/v6/cpu/748661
    
    YABS completed in 7 min 31 sec
    
    

    CPX11

    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    #              Yet-Another-Bench-Script              #
    #                     v2023-03-24                    #
    # https://github.com/masonr/yet-another-bench-script #
    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    
    Fri Mar 31 01:34:42 PM UTC 2023
    
    Basic System Information:
    ---------------------------------
    Uptime     : 6 days, 38 hours, 0 minutes
    Processor  : AMD EPYC Processor
    CPU cores  : 2 @ 2495.286 MHz
    AES-NI     : ✔ Enabled
    VM-x/AMD-V : ❌ Disabled
    RAM        : 1.9 GiB
    Swap       : 0.0 KiB
    Disk       : 37.5 GiB
    Distro     : Ubuntu 22.04.2 LTS
    Kernel     : 5.15.0-67-generic
    VM Type    : KVM
    Net Online : IPv4 & IPv6
    
    IPv6 Network Information:
    ---------------------------------
    ISP        : Hetzner Online GmbH
    ASN        : AS24940 Hetzner Online GmbH
    Host       : Hetzner
    Location   : Nuremberg, Bavaria (BY)
    Country    : Germany
    
    fio Disk Speed Tests (Mixed R/W 50/50):
    ---------------------------------
    Block Size | 4k            (IOPS) | 64k           (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 140.70 MB/s  (35.1k) | 1.35 GB/s    (21.2k)
    Write      | 141.07 MB/s  (35.2k) | 1.36 GB/s    (21.3k)
    Total      | 281.77 MB/s  (70.4k) | 2.72 GB/s    (42.5k)
               |                      |                     
    Block Size | 512k          (IOPS) | 1m            (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 1.64 GB/s     (3.2k) | 1.61 GB/s     (1.5k)
    Write      | 1.73 GB/s     (3.3k) | 1.71 GB/s     (1.6k)
    Total      | 3.37 GB/s     (6.6k) | 3.32 GB/s     (3.2k)
    
    iperf3 Network Speed Tests (IPv4):
    ---------------------------------
    Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping           
    -----           | -----                     | ----            | ----            | ----           
    Clouvider       | London, UK (10G)          | 7.01 Gbits/sec  | 8.52 Gbits/sec  | 20.9 ms        
    Scaleway        | Paris, FR (10G)           | busy            | 7.89 Gbits/sec  | 13.4 ms        
    NovoServe       | North Holland, NL (40G)   | 9.76 Gbits/sec  | 11.4 Gbits/sec  | 10.5 ms        
    Uztelecom       | Tashkent, UZ (10G)        | 2.31 Gbits/sec  | 2.11 Gbits/sec  | 85.5 ms        
    Clouvider       | NYC, NY, US (10G)         | 762 Mbits/sec   | 165 Mbits/sec   | 83.3 ms        
    Clouvider       | Dallas, TX, US (10G)      | 1.45 Gbits/sec  | 1.50 Gbits/sec  | 124 ms         
    Clouvider       | Los Angeles, CA, US (10G) | 961 Mbits/sec   | 1.22 Gbits/sec  | 144 ms         
    
    iperf3 Network Speed Tests (IPv6):
    ---------------------------------
    Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping           
    -----           | -----                     | ----            | ----            | ----           
    Clouvider       | London, UK (10G)          | busy            | 4.53 Gbits/sec  | 20.7 ms        
    Scaleway        | Paris, FR (10G)           | busy            | 8.29 Gbits/sec  | 13.6 ms        
    NovoServe       | North Holland, NL (40G)   | busy            | 10.6 Gbits/sec  | 10.3 ms        
    Uztelecom       | Tashkent, UZ (10G)        | 2.14 Gbits/sec  | 1.92 Gbits/sec  | 85.4 ms        
    Clouvider       | NYC, NY, US (10G)         | busy            | 523 Mbits/sec   | 83.2 ms        
    Clouvider       | Dallas, TX, US (10G)      | 1.25 Gbits/sec  | 1.49 Gbits/sec  | 124 ms         
    Clouvider       | Los Angeles, CA, US (10G) | 1.01 Gbits/sec  | 1.23 Gbits/sec  | 144 ms         
    
    Geekbench 6 test failed. Run manually to determine cause.
    
    YABS completed in 9 min 56 sec
    
    

    CPX51

    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    #              Yet-Another-Bench-Script              #
    #                     v2023-03-24                    #
    # https://github.com/masonr/yet-another-bench-script #
    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    
    Fri Mar 31 01:38:53 PM UTC 2023
    
    Basic System Information:
    ---------------------------------
    Uptime     : 0 days, 0 hours, 2 minutes
    Processor  : AMD EPYC Processor
    CPU cores  : 16 @ 2445.404 MHz
    AES-NI     : ✔ Enabled
    VM-x/AMD-V : ❌ Disabled
    RAM        : 30.6 GiB
    Swap       : 0.0 KiB
    Disk       : 338.0 GiB
    Distro     : Ubuntu 22.04.2 LTS
    Kernel     : 5.15.0-67-generic
    VM Type    : KVM
    Net Online : IPv4 & IPv6
    
    IPv6 Network Information:
    ---------------------------------
    ISP        : Hetzner Online GmbH
    ASN        : AS24940 Hetzner Online GmbH
    Host       : Hetzner Online GmbH
    Location   : Tuusula, Uusimaa (18)
    Country    : Finland
    
    fio Disk Speed Tests (Mixed R/W 50/50):
    ---------------------------------
    Block Size | 4k            (IOPS) | 64k           (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 162.19 MB/s  (40.5k) | 1.81 GB/s    (28.3k)
    Write      | 162.62 MB/s  (40.6k) | 1.82 GB/s    (28.5k)
    Total      | 324.81 MB/s  (81.2k) | 3.64 GB/s    (56.9k)
               |                      |                     
    Block Size | 512k          (IOPS) | 1m            (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 3.00 GB/s     (5.8k) | 3.31 GB/s     (3.2k)
    Write      | 3.16 GB/s     (6.1k) | 3.53 GB/s     (3.4k)
    Total      | 6.17 GB/s    (12.0k) | 6.85 GB/s     (6.6k)
    
    iperf3 Network Speed Tests (IPv4):
    ---------------------------------
    Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping           
    -----           | -----                     | ----            | ----            | ----           
    Clouvider       | London, UK (10G)          | 3.39 Gbits/sec  | busy            | 33.7 ms        
    Scaleway        | Paris, FR (10G)           | 6.06 Gbits/sec  | 4.55 Gbits/sec  | 36.6 ms        
    NovoServe       | North Holland, NL (40G)   | 6.63 Gbits/sec  | busy            | 28.3 ms        
    Uztelecom       | Tashkent, UZ (10G)        | busy            | 3.00 Gbits/sec  | 57.1 ms        
    Clouvider       | NYC, NY, US (10G)         | 978 Mbits/sec   | 415 Mbits/sec   | 101 ms         
    Clouvider       | Dallas, TX, US (10G)      | 971 Mbits/sec   | 1.28 Gbits/sec  | 134 ms         
    Clouvider       | Los Angeles, CA, US (10G) | 923 Mbits/sec   | 1.03 Gbits/sec  | 158 ms         
    
    iperf3 Network Speed Tests (IPv6):
    ---------------------------------
    Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping           
    -----           | -----                     | ----            | ----            | ----           
    Clouvider       | London, UK (10G)          | 4.17 Gbits/sec  | 3.11 Gbits/sec  | 33.4 ms        
    Scaleway        | Paris, FR (10G)           | 5.38 Gbits/sec  | 5.23 Gbits/sec  | 37.8 ms        
    NovoServe       | North Holland, NL (40G)   | 6.55 Gbits/sec  | 6.96 Gbits/sec  | 28.4 ms        
    Uztelecom       | Tashkent, UZ (10G)        | 3.02 Gbits/sec  | 2.95 Gbits/sec  | 57.0 ms        
    Clouvider       | NYC, NY, US (10G)         | 1.42 Gbits/sec  | 1.78 Gbits/sec  | 101 ms         
    Clouvider       | Dallas, TX, US (10G)      | 1.18 Gbits/sec  | 1.37 Gbits/sec  | 134 ms         
    Clouvider       | Los Angeles, CA, US (10G) | 1.03 Gbits/sec  | 1.13 Gbits/sec  | 158 ms         
    
    Geekbench 6 Benchmark Test:
    ---------------------------------
    Test            | Value                         
                    |                               
    Single Core     | 1349                          
    Multi Core      | 9794                          
    Full Test       | https://browser.geekbench.com/v6/cpu/748741
    
    YABS completed in 10 min 41 sec
    
    

    For comparison, our MD3 with 2TB NVMe (40€ a month before discounts):

    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    #              Yet-Another-Bench-Script              #
    #                     v2023-02-27                    #
    # https://github.com/masonr/yet-another-bench-script #
    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    
    Fri 03 Mar 2023 08:31:58 PM EET
    
    Basic System Information:
    ---------------------------------
    Uptime     : 10 days, 3 hours, 3 minutes
    Processor  : Intel(R) Core(TM) i7-7700T CPU @ 2.90GHz
    CPU cores  : 8 @ 3191.241 MHz
    AES-NI     : ✔ Enabled
    VM-x/AMD-V : ✔ Enabled
    RAM        : 31.2 GiB
    Swap       : 2.3 GiB
    Disk       : 1.8 TiB
    Distro     : Debian GNU/Linux 11 (bullseye)
    Kernel     : 5.10.0-18-amd64
    VM Type    : NONE
    
    Basic Network Information:
    ---------------------------------
    Protocol   : IPv4
    ISP        : Magna Capax Finland Oy
    ASN        : AS203003 Magna Capax Finland Oy
    Host       : Magnacapax
    Location   : Helsinki, Uusimaa (18)
    Country    : Finland
    
    fio Disk Speed Tests (Mixed R/W 50/50):
    ---------------------------------
    Block Size | 4k            (IOPS) | 64k           (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 462.13 MB/s (115.5k) | 642.39 MB/s  (10.0k)
    Write      | 463.35 MB/s (115.8k) | 645.77 MB/s  (10.0k)
    Total      | 925.48 MB/s (231.3k) | 1.28 GB/s    (20.1k)
               |                      |                     
    Block Size | 512k          (IOPS) | 1m            (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 838.62 MB/s   (1.6k) | 1.02 GB/s     (1.0k)
    Write      | 883.17 MB/s   (1.7k) | 1.09 GB/s     (1.0k)
    Total      | 1.72 GB/s     (3.3k) | 2.11 GB/s     (2.0k)
    
    iperf3 Network Speed Tests (IPv4):
    ---------------------------------
    Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping           
    -----           | -----                     | ----            | ----            | ----           
    Clouvider       | London, UK (10G)          | 819 Mbits/sec   | 927 Mbits/sec   | 36.3 ms        
    Scaleway        | Paris, FR (10G)           | 828 Mbits/sec   | busy            | 34.5 ms        
    NovoServe       | North Holland, NL (40G)   | 823 Mbits/sec   | 930 Mbits/sec   | 29.4 ms        
    Uztelecom       | Tashkent, UZ (10G)        | 782 Mbits/sec   | 881 Mbits/sec   | 110 ms         
    Clouvider       | NYC, NY, US (10G)         | 786 Mbits/sec   | 893 Mbits/sec   | 91.3 ms        
    Clouvider       | Dallas, TX, US (10G)      | 758 Mbits/sec   | 862 Mbits/sec   | 136 ms         
    Clouvider       | Los Angeles, CA, US (10G) | 142 Mbits/sec   | 493 Mbits/sec   | 166 ms         
    
    Geekbench 6 Benchmark Test:
    ---------------------------------
    Test            | Value                         
                    |                               
    Single Core     | 1448                          
    Multi Core      | 4697                          
    Full Test       | https://browser.geekbench.com/v6/cpu/395087
    
    YABS completed in 7 min 49 sec
    
    
    Thanked by 3maverick JamesF ralf
  • nullroutenullroute Member, Host Rep

    @EthanZou said:

    @nullroute said: 40G port;

    Not 10 Gbit/s-Port?

    No, according to some members of the technical team it is a 40g port with redundancy.

  • @PulsedMedia said:
    Yabs results:

    AX51-NVME (Raid1)

    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    #              Yet-Another-Bench-Script              #
    #                     v2023-03-24                    #
    # https://github.com/masonr/yet-another-bench-script #
    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    
    Fri 31 Mar 2023 03:34:38 PM CEST
    
    Basic System Information:
    ---------------------------------
    Uptime     : 186 days, 5 hours, 41 minutes
    Processor  : AMD Ryzen 7 3700X 8-Core Processor
    CPU cores  : 16 @ 2139.387 MHz
    AES-NI     : ✔ Enabled
    VM-x/AMD-V : ✔ Enabled
    RAM        : 62.8 GiB
    Swap       : 32.0 GiB
    Disk       : 906.2 GiB
    Distro     : Debian GNU/Linux 11 (bullseye)
    Kernel     : 5.10.0-18-amd64
    VM Type    : NONE
    Net Online : IPv4
    
    IPv4 Network Information:
    ---------------------------------
    ISP        : Hetzner Online GmbH
    ASN        : AS24940 Hetzner Online GmbH
    Host       : Hetzner
    Location   : Helsinki, Uusimaa (18)
    Country    : Finland
    
    fio Disk Speed Tests (Mixed R/W 50/50):
    ---------------------------------
    Block Size | 4k            (IOPS) | 64k           (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 625.93 MB/s (156.4k) | 1.21 GB/s    (18.9k)
    Write      | 627.58 MB/s (156.8k) | 1.21 GB/s    (19.0k)
    Total      | 1.25 GB/s   (313.3k) | 2.42 GB/s    (37.9k)
               |                      |                     
    Block Size | 512k          (IOPS) | 1m            (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 971.87 MB/s   (1.8k) | 1.01 GB/s      (991)
    Write      | 1.02 GB/s     (1.9k) | 1.08 GB/s     (1.0k)
    Total      | 1.99 GB/s     (3.8k) | 2.09 GB/s     (2.0k)
    
    iperf3 Network Speed Tests (IPv4):
    ---------------------------------
    Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping           
    -----           | -----                     | ----            | ----            | ----           
    Clouvider       | London, UK (10G)          | 934 Mbits/sec   | 259 Mbits/sec   | 36.3 ms        
    Scaleway        | Paris, FR (10G)           | busy            | 104 Mbits/sec   | 36.4 ms        
    NovoServe       | North Holland, NL (40G)   | 937 Mbits/sec   | busy            | 30.2 ms        
    Uztelecom       | Tashkent, UZ (10G)        | 915 Mbits/sec   | 119 Mbits/sec   | 60.8 ms        
    Clouvider       | NYC, NY, US (10G)         | 791 Mbits/sec   | 98.4 Mbits/sec  | 97.4 ms        
    Clouvider       | Dallas, TX, US (10G)      | 536 Mbits/sec   | 44.9 Mbits/sec  | 139 ms         
    Clouvider       | Los Angeles, CA, US (10G) | 175 Mbits/sec   | 21.3 Mbits/sec  | 170 ms         
    
    Geekbench 6 Benchmark Test:
    ---------------------------------
    Test            | Value                         
                    |                               
    Single Core     | 1717                          
    Multi Core      | 8641                          
    Full Test       | https://browser.geekbench.com/v6/cpu/748661
    
    YABS completed in 7 min 31 sec
    
    

    CPX11

    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    #              Yet-Another-Bench-Script              #
    #                     v2023-03-24                    #
    # https://github.com/masonr/yet-another-bench-script #
    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    
    Fri Mar 31 01:34:42 PM UTC 2023
    
    Basic System Information:
    ---------------------------------
    Uptime     : 6 days, 38 hours, 0 minutes
    Processor  : AMD EPYC Processor
    CPU cores  : 2 @ 2495.286 MHz
    AES-NI     : ✔ Enabled
    VM-x/AMD-V : ❌ Disabled
    RAM        : 1.9 GiB
    Swap       : 0.0 KiB
    Disk       : 37.5 GiB
    Distro     : Ubuntu 22.04.2 LTS
    Kernel     : 5.15.0-67-generic
    VM Type    : KVM
    Net Online : IPv4 & IPv6
    
    IPv6 Network Information:
    ---------------------------------
    ISP        : Hetzner Online GmbH
    ASN        : AS24940 Hetzner Online GmbH
    Host       : Hetzner
    Location   : Nuremberg, Bavaria (BY)
    Country    : Germany
    
    fio Disk Speed Tests (Mixed R/W 50/50):
    ---------------------------------
    Block Size | 4k            (IOPS) | 64k           (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 140.70 MB/s  (35.1k) | 1.35 GB/s    (21.2k)
    Write      | 141.07 MB/s  (35.2k) | 1.36 GB/s    (21.3k)
    Total      | 281.77 MB/s  (70.4k) | 2.72 GB/s    (42.5k)
               |                      |                     
    Block Size | 512k          (IOPS) | 1m            (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 1.64 GB/s     (3.2k) | 1.61 GB/s     (1.5k)
    Write      | 1.73 GB/s     (3.3k) | 1.71 GB/s     (1.6k)
    Total      | 3.37 GB/s     (6.6k) | 3.32 GB/s     (3.2k)
    
    iperf3 Network Speed Tests (IPv4):
    ---------------------------------
    Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping           
    -----           | -----                     | ----            | ----            | ----           
    Clouvider       | London, UK (10G)          | 7.01 Gbits/sec  | 8.52 Gbits/sec  | 20.9 ms        
    Scaleway        | Paris, FR (10G)           | busy            | 7.89 Gbits/sec  | 13.4 ms        
    NovoServe       | North Holland, NL (40G)   | 9.76 Gbits/sec  | 11.4 Gbits/sec  | 10.5 ms        
    Uztelecom       | Tashkent, UZ (10G)        | 2.31 Gbits/sec  | 2.11 Gbits/sec  | 85.5 ms        
    Clouvider       | NYC, NY, US (10G)         | 762 Mbits/sec   | 165 Mbits/sec   | 83.3 ms        
    Clouvider       | Dallas, TX, US (10G)      | 1.45 Gbits/sec  | 1.50 Gbits/sec  | 124 ms         
    Clouvider       | Los Angeles, CA, US (10G) | 961 Mbits/sec   | 1.22 Gbits/sec  | 144 ms         
    
    iperf3 Network Speed Tests (IPv6):
    ---------------------------------
    Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping           
    -----           | -----                     | ----            | ----            | ----           
    Clouvider       | London, UK (10G)          | busy            | 4.53 Gbits/sec  | 20.7 ms        
    Scaleway        | Paris, FR (10G)           | busy            | 8.29 Gbits/sec  | 13.6 ms        
    NovoServe       | North Holland, NL (40G)   | busy            | 10.6 Gbits/sec  | 10.3 ms        
    Uztelecom       | Tashkent, UZ (10G)        | 2.14 Gbits/sec  | 1.92 Gbits/sec  | 85.4 ms        
    Clouvider       | NYC, NY, US (10G)         | busy            | 523 Mbits/sec   | 83.2 ms        
    Clouvider       | Dallas, TX, US (10G)      | 1.25 Gbits/sec  | 1.49 Gbits/sec  | 124 ms         
    Clouvider       | Los Angeles, CA, US (10G) | 1.01 Gbits/sec  | 1.23 Gbits/sec  | 144 ms         
    
    Geekbench 6 test failed. Run manually to determine cause.
    
    YABS completed in 9 min 56 sec
    
    

    CPX51

    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    #              Yet-Another-Bench-Script              #
    #                     v2023-03-24                    #
    # https://github.com/masonr/yet-another-bench-script #
    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    
    Fri Mar 31 01:38:53 PM UTC 2023
    
    Basic System Information:
    ---------------------------------
    Uptime     : 0 days, 0 hours, 2 minutes
    Processor  : AMD EPYC Processor
    CPU cores  : 16 @ 2445.404 MHz
    AES-NI     : ✔ Enabled
    VM-x/AMD-V : ❌ Disabled
    RAM        : 30.6 GiB
    Swap       : 0.0 KiB
    Disk       : 338.0 GiB
    Distro     : Ubuntu 22.04.2 LTS
    Kernel     : 5.15.0-67-generic
    VM Type    : KVM
    Net Online : IPv4 & IPv6
    
    IPv6 Network Information:
    ---------------------------------
    ISP        : Hetzner Online GmbH
    ASN        : AS24940 Hetzner Online GmbH
    Host       : Hetzner Online GmbH
    Location   : Tuusula, Uusimaa (18)
    Country    : Finland
    
    fio Disk Speed Tests (Mixed R/W 50/50):
    ---------------------------------
    Block Size | 4k            (IOPS) | 64k           (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 162.19 MB/s  (40.5k) | 1.81 GB/s    (28.3k)
    Write      | 162.62 MB/s  (40.6k) | 1.82 GB/s    (28.5k)
    Total      | 324.81 MB/s  (81.2k) | 3.64 GB/s    (56.9k)
               |                      |                     
    Block Size | 512k          (IOPS) | 1m            (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 3.00 GB/s     (5.8k) | 3.31 GB/s     (3.2k)
    Write      | 3.16 GB/s     (6.1k) | 3.53 GB/s     (3.4k)
    Total      | 6.17 GB/s    (12.0k) | 6.85 GB/s     (6.6k)
    
    iperf3 Network Speed Tests (IPv4):
    ---------------------------------
    Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping           
    -----           | -----                     | ----            | ----            | ----           
    Clouvider       | London, UK (10G)          | 3.39 Gbits/sec  | busy            | 33.7 ms        
    Scaleway        | Paris, FR (10G)           | 6.06 Gbits/sec  | 4.55 Gbits/sec  | 36.6 ms        
    NovoServe       | North Holland, NL (40G)   | 6.63 Gbits/sec  | busy            | 28.3 ms        
    Uztelecom       | Tashkent, UZ (10G)        | busy            | 3.00 Gbits/sec  | 57.1 ms        
    Clouvider       | NYC, NY, US (10G)         | 978 Mbits/sec   | 415 Mbits/sec   | 101 ms         
    Clouvider       | Dallas, TX, US (10G)      | 971 Mbits/sec   | 1.28 Gbits/sec  | 134 ms         
    Clouvider       | Los Angeles, CA, US (10G) | 923 Mbits/sec   | 1.03 Gbits/sec  | 158 ms         
    
    iperf3 Network Speed Tests (IPv6):
    ---------------------------------
    Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping           
    -----           | -----                     | ----            | ----            | ----           
    Clouvider       | London, UK (10G)          | 4.17 Gbits/sec  | 3.11 Gbits/sec  | 33.4 ms        
    Scaleway        | Paris, FR (10G)           | 5.38 Gbits/sec  | 5.23 Gbits/sec  | 37.8 ms        
    NovoServe       | North Holland, NL (40G)   | 6.55 Gbits/sec  | 6.96 Gbits/sec  | 28.4 ms        
    Uztelecom       | Tashkent, UZ (10G)        | 3.02 Gbits/sec  | 2.95 Gbits/sec  | 57.0 ms        
    Clouvider       | NYC, NY, US (10G)         | 1.42 Gbits/sec  | 1.78 Gbits/sec  | 101 ms         
    Clouvider       | Dallas, TX, US (10G)      | 1.18 Gbits/sec  | 1.37 Gbits/sec  | 134 ms         
    Clouvider       | Los Angeles, CA, US (10G) | 1.03 Gbits/sec  | 1.13 Gbits/sec  | 158 ms         
    
    Geekbench 6 Benchmark Test:
    ---------------------------------
    Test            | Value                         
                    |                               
    Single Core     | 1349                          
    Multi Core      | 9794                          
    Full Test       | https://browser.geekbench.com/v6/cpu/748741
    
    YABS completed in 10 min 41 sec
    
    

    For comparison, our MD3 with 2TB NVMe (40€ a month before discounts):

    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    #              Yet-Another-Bench-Script              #
    #                     v2023-02-27                    #
    # https://github.com/masonr/yet-another-bench-script #
    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    
    Fri 03 Mar 2023 08:31:58 PM EET
    
    Basic System Information:
    ---------------------------------
    Uptime     : 10 days, 3 hours, 3 minutes
    Processor  : Intel(R) Core(TM) i7-7700T CPU @ 2.90GHz
    CPU cores  : 8 @ 3191.241 MHz
    AES-NI     : ✔ Enabled
    VM-x/AMD-V : ✔ Enabled
    RAM        : 31.2 GiB
    Swap       : 2.3 GiB
    Disk       : 1.8 TiB
    Distro     : Debian GNU/Linux 11 (bullseye)
    Kernel     : 5.10.0-18-amd64
    VM Type    : NONE
    
    Basic Network Information:
    ---------------------------------
    Protocol   : IPv4
    ISP        : Magna Capax Finland Oy
    ASN        : AS203003 Magna Capax Finland Oy
    Host       : Magnacapax
    Location   : Helsinki, Uusimaa (18)
    Country    : Finland
    
    fio Disk Speed Tests (Mixed R/W 50/50):
    ---------------------------------
    Block Size | 4k            (IOPS) | 64k           (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 462.13 MB/s (115.5k) | 642.39 MB/s  (10.0k)
    Write      | 463.35 MB/s (115.8k) | 645.77 MB/s  (10.0k)
    Total      | 925.48 MB/s (231.3k) | 1.28 GB/s    (20.1k)
               |                      |                     
    Block Size | 512k          (IOPS) | 1m            (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 838.62 MB/s   (1.6k) | 1.02 GB/s     (1.0k)
    Write      | 883.17 MB/s   (1.7k) | 1.09 GB/s     (1.0k)
    Total      | 1.72 GB/s     (3.3k) | 2.11 GB/s     (2.0k)
    
    iperf3 Network Speed Tests (IPv4):
    ---------------------------------
    Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping           
    -----           | -----                     | ----            | ----            | ----           
    Clouvider       | London, UK (10G)          | 819 Mbits/sec   | 927 Mbits/sec   | 36.3 ms        
    Scaleway        | Paris, FR (10G)           | 828 Mbits/sec   | busy            | 34.5 ms        
    NovoServe       | North Holland, NL (40G)   | 823 Mbits/sec   | 930 Mbits/sec   | 29.4 ms        
    Uztelecom       | Tashkent, UZ (10G)        | 782 Mbits/sec   | 881 Mbits/sec   | 110 ms         
    Clouvider       | NYC, NY, US (10G)         | 786 Mbits/sec   | 893 Mbits/sec   | 91.3 ms        
    Clouvider       | Dallas, TX, US (10G)      | 758 Mbits/sec   | 862 Mbits/sec   | 136 ms         
    Clouvider       | Los Angeles, CA, US (10G) | 142 Mbits/sec   | 493 Mbits/sec   | 166 ms         
    
    Geekbench 6 Benchmark Test:
    ---------------------------------
    Test            | Value                         
                    |                               
    Single Core     | 1448                          
    Multi Core      | 4697                          
    Full Test       | https://browser.geekbench.com/v6/cpu/395087
    
    YABS completed in 7 min 49 sec
    
    

    Good job! Cloud yes!

    @nullroute said:

    @EthanZou said:

    @nullroute said: 40G port;

    Not 10 Gbit/s-Port?

    No, according to some members of the technical team it is a 40g port with redundancy.

    That's great! B)

    Thanked by 1JamesF
  • King Kong
    AMD Ryzen™ 9 5950X
    128 GB DDR4 ECC
    2 x 3.84 TB NVMe SSD
    1 Gbit/s-Port Guaranteed
    € 103.30 monthly

  • Just the Raid, the 10gbps network + snapshots option are enough to justify the +12€ difference imho

  • @qquccs said:
    Just the Raid, the 10gbps network + snapshots option are enough to justify the +12€ difference imho

    Yes, and DDoS protection.

  • JamesFJamesF Member, Host Rep

    Is the hetzner cloud better than the OVH VPS?

    seems hetzner don't offer cpanel for cloud either

  • @JamesF said: Is the hetzner cloud better than the OVH VPS?

    Yes, maybe, but Hetzner is much more cost-effective.

    Thanked by 1JamesF
  • iwaswrongonceiwaswrongonce Member
    edited April 2023

    No clue what you guys are on about. Shared cloud maxes out at 32gb. The AX51 has 64gb. 5x better 4k iops. Way better CPU in practice (much better single thread, no noisy neighbors).

    But it gets insane if you're actually deploying anything more than a toy 32gb instance. AX101 is 109 EUR in Germany and the sorta comparable CCX52 is 296 EUR.

  • PulsedMediaPulsedMedia Member, Patron Provider

    @iwaswrongonce said: 5x better 4k iops. Way better CPU in practice (much better single thread, no noisy neighbors).

    indeed and the average task the 4k iops is most important :) say webhosting, database etc. only large file stuff benefits from the bigger block sizes.

  • @iwaswrongonce said: a toy 32gb instance

    Toy?

    AX101 is 109 EUR in Germany and the sorta comparable CCX52 is 296 EUR.

    That's amazing!

    @PulsedMedia said: indeed and the average task the 4k iops is most important say webhosting, database etc. only large file stuff benefits from the bigger block sizes.

    So which one do you prefer? dedi or cloud?

  • For us it’s the hourly billing.

    We fire up a 48 core Cloud server to compile our softwares and using multi threading, compiling gets done in mere seconds. We use that server 1-hour max every month. That’s less than €1.

    I can’t think of any other way to do this that is faster and cheaper. You deploy a 48 core server in a few seconds, compile your software on it blazingly fast for pennies then be done with the server.

    I don’t see a reason to use such server for a full month though. It’ll cost ~€400. It’s better to get a dedicated for a full-time workload at that point.

    Also everything Hetzner Cloud provides; Networks, Load Balancers, Failover IPs, Firewalls are available for their dedicated servers as well. Storage Boxes can be used for backups.

    So there’s a use case for everything. If those 48 core Hetzner Cloud servers weren’t exist, we would be paying way more for some other solution.

  • @anbelevebel said: I don’t see a reason to use such server for a full month though. It’ll cost ~€400. It’s better to get a dedicated for a full-time workload at that point.

    Agree, but there is no Hetzner dedi in the US.
    So any suggestions? thanks!

  • ralfralf Member

    @anbelevebel said:
    For us it’s the hourly billing.

    We fire up a 48 core Cloud server to compile our softwares and using multi threading, compiling gets done in mere seconds. We use that server 1-hour max every month. That’s less than €1.

    I don’t see a reason to use such server for a full month though. It’ll cost ~€400. It’s better to get a dedicated for a full-time workload at that point.

    So do you only compile once per month? I've not used the Hetzner cloud yet, only a dedi, but everything I've said (and what you said above) says hourly billing, not per-minute or per-second.

    I'm sure that being charged €1 per hour for only a few times is also worth it, but the way you've written it above strongly suggests you only use a couple of minutes at a time, and cumulatively it's less than an hour and only billed for 1 hour across the whole month.

    Actually, I just checked https://docs.hetzner.com/cloud/billing/faq

    Mathematically speaking, we will bill you for the minimum amount, whether that is the monthly price cap OR the hourly price multiplied by the number of hours you used the server.
    We always round up the hourly usage of a server. If you create a server just for a few minutes, we will still bill you for one whole hour. Why? When you created the server, it placed an additional load on our systems.

  • @ralf said:

    @anbelevebel said:
    For us it’s the hourly billing.

    We fire up a 48 core Cloud server to compile our softwares and using multi threading, compiling gets done in mere seconds. We use that server 1-hour max every month. That’s less than €1.

    I don’t see a reason to use such server for a full month though. It’ll cost ~€400. It’s better to get a dedicated for a full-time workload at that point.

    So do you only compile once per month? I've not used the Hetzner cloud yet, only a dedi, but everything I've said (and what you said above) says hourly billing, not per-minute or per-second.

    I'm sure that being charged €1 per hour for only a few times is also worth it, but the way you've written it above strongly suggests you only use a couple of minutes at a time, and cumulatively it's less than an hour and only billed for 1 hour across the whole month.

    Actually, I just checked https://docs.hetzner.com/cloud/billing/faq

    Mathematically speaking, we will bill you for the minimum amount, whether that is the monthly price cap OR the hourly price multiplied by the number of hours you used the server.
    We always round up the hourly usage of a server. If you create a server just for a few minutes, we will still bill you for one whole hour. Why? When you created the server, it placed an additional load on our systems.

    Sorry my mistake. We compile a few times a month. It should be about €5 or €6 a month tops.

    Thanked by 1ralf
  • ralfralf Member

    @ralf said:
    ... but everything I've said ...

    (typo) everything i've read

  • @stefeman said: High-Availability is what costs in the cloud.

    To be fair Hetzner Cloud isn't a HA environment. Its just a VPS with more features

    If the server your VM on goes kaput then you're out of luck until they fix it, unless you have a second VM behind a load balancer (they are HA) which can take the load.

    If you do that then use placement groups to ensure they aren't on the same physical host of course

  • emghemgh Member
    edited April 2023

    @lukehebb said:

    @stefeman said: High-Availability is what costs in the cloud.

    To be fair Hetzner Cloud isn't a HA environment. Its just a VPS with more features

    If the server your VM on goes kaput then you're out of luck until they fix it, unless you have a second VM behind a load balancer (they are HA) which can take the load.

    If you do that then use placement groups to ensure they aren't on the same physical host of course

    Kaput isn’t really a technical term so it’s hard to argue your stance.

    However, for example, in some (maybe limited) instances, if the server isn’t booting and it’s not the drive, I doubt their backend don’t move your instance to another node within the hypervisor.

  • @emgh said:
    move your instance to another node within the hypervisor.

    You understand the node is the hypervisor yes?

  • emghemgh Member

    @iwaswrongonce said:

    @emgh said:
    move your instance to another node within the hypervisor.

    You understand the node is the hypervisor yes?

    You're right, across nodes & hypervisors. I'm tired.

    My point is, I would believe their system is self-healing to some extent.

  • @emgh said:
    My point is, I would believe their system is self-healing to some extent.

    I think you would be surprised.

  • emghemgh Member
    edited April 2023

    @iwaswrongonce said:

    @emgh said:
    My point is, I would believe their system is self-healing to some extent.

    I think you would be surprised.

    You might be right.

    They locked me out of my account because of their actions (not charging me in time), and not allowing me to manually pay because of the ongoing charge that they're themselves still processing.

    I save the error messages.

    The payment gets processed, I get unlocked.

    I contact support, they say I wasn't locked out because it's not in their logs.

    I give them the exact error messages I was seeing, they say it didn't happen because it's not in their logs.

  • @lukehebb said:

    @stefeman said: High-Availability is what costs in the cloud.

    To be fair Hetzner Cloud isn't a HA environment. Its just a VPS with more features

    If the server your VM on goes kaput then you're out of luck until they fix it, unless you have a second VM behind a load balancer (they are HA) which can take the load.

    If you do that then use placement groups to ensure they aren't on the same physical host of course

    @iwaswrongonce said:

    @emgh said:
    move your instance to another node within the hypervisor.

    You understand the node is the hypervisor yes?

    Not Cloud, but VPS? @Hetzner_OL

  • @lukehebb said:

    @stefeman said: High-Availability is what costs in the cloud.

    To be fair Hetzner Cloud isn't a HA environment. Its just a VPS with more features

    If the server your VM on goes kaput then you're out of luck until they fix it, unless you have a second VM behind a load balancer (they are HA) which can take the load.

    If you do that then use placement groups to ensure they aren't on the same physical host of course

    I dont know about ha, but are they using openstack with software defined storage, vrack and all those scalable infra?

  • davidedavide Member
    edited April 2023

    Hetzner uses consumer computers for the dedi plans to offer competitive prices. Their cloud offers use real servers instead. Their income tax bracket in Germany should be ~50% so necessarily they need to cut corners to appear competitive. I used consumer computers for 20 years, but since I bought a xeon supermicro, the improvement in stability and reliability has been day and night compared to the consumer crap I previously used.

  • @davide said:
    Hetzner uses consumer computers for the dedi plans to offer competitive prices. Their cloud offers use real servers instead. Their income tax bracket in Germany should be ~50% so necessarily they need to cut corners to appear competitive. I used consumer computers for 20 years, but since I bought a xeon supermicro, the improvement in stability and reliability has been day and night compared to the consumer crap I previously used.

    But the Yabs results of Hetzner dedis look great, even better than their clouds.

Sign In or Register to comment.