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

› › Hetzner CCX31 vs Netcup RS 8000 G9 – Dedicated vCPU 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.

Hetzner CCX31 vs Netcup RS 8000 G9 – Dedicated vCPU Speed?

nqservicesnqservices Member
in General

Hi,

I need a powerful VPS to host some Wordpress websites, and I’m trying to decide if I buy the Hetzner CCX31 or the Netcup RS 8000 G9. I narrow my search to this two, since I need it to be hosted in Europe and to have Dedicated vCPU.

Both say on their websites that they have Dedicated vCPU, and for WordPress a fast CPU is important.

Netcup price is lower and has a much bigger disk, more 2 cores (10) and double RAM (64GB) when comparing with Hetzner. Still Hetzner is more expensive. Disk size does not matter for me, since 100GB will meet my requirements.

My priority is to buy the one with best CPU performance to host Wordpress.

Any advice is appreciated. If someone has one of these, posting a bench would be great!

Thanks

Comments

  • LeviLevi Member

    When in doubt always choose hatznar. Netcup has harsh penalty for questionable billing practices.

    Also, consider contabo.

  • SpeedTestSpeedTest Member

    I would go with AX51-NVMe or AX61-NVMe for Dedicated CPU. And vServer XL SSD G3 AMD from php-friends.de if it has to be VPS

  • AC_FanAC_Fan Member

    Hetzner CCX series doesn't make much sense to me. What's your actual requirements for WordPress (I'm assuming you have multiple WP sites, since you're looking for dedicated CPU)? Overall, I would say the AX41-NVMe might be your best bet, although NetCup, PHP-Friends and Avoro are also all valid options.

  • mezoologymezoology Member
    edited November 2020

    how to find out that you have a really dedicated CPU, if you happen to rent one?

    also

    @LTniger said:
    When in doubt always choose hatznar. Netcup has harsh penalty for questionable billing practices.

    care to elaborate more ?

  • mezoologymezoology Member
    edited November 2020

    how to find out that you have a really dedicated CPU, if you happen to rent one?
    also

    @LTniger said:
    Netcup has harsh penalty for questionable billing practices.

    care to elaborate more ?

    edit: sorry my internet shat itself.

  • ZerpyZerpy Member
    edited November 2020

    @LTniger said:
    When in doubt always choose hatznar. Netcup has harsh penalty for questionable billing practices.

    Not really - can't see how german billing practices are questionable - obviously that you fail to understand them, isn't a netcup problem, but rather a you problem.

    ^OP:
    I'd go for Netcup - they have overall better performance than hetzner's dedicated VMs - also works out cheaper anyway.

  • sonicsonic Veteran

    @Zerpy said:

    @LTniger said:
    When in doubt always choose hatznar. Netcup has harsh penalty for questionable billing practices.

    Not really - can't see how german billing practices are questionable - obviously that you fail to understand them, isn't a netcup problem, but rather a you problem.

    ^OP:
    I'd go for Netcup - they have overall better performance than hetzner's dedicated VMs - also works out cheaper anyway.

    How about Netcup uptime? Are their network solid?
    I couldnt find their status page to see uptime events.

  • hacktekhacktek Member

    Netcup hosts in hetzner DCs

  • LeviLevi Member

    @Zerpy said:

    @LTniger said:
    When in doubt always choose hatznar. Netcup has harsh penalty for questionable billing practices.

    Not really - can't see how german billing practices are questionable - obviously that you fail to understand them, isn't a netcup problem, but rather a you problem.

    If hetzner can do normal billing, than it is not german specific. Just provider is a dick. How much dick? Use search.

  • ZerpyZerpy Member

    @sonic said:
    How about Netcup uptime? Are their network solid?
    I couldnt find their status page to see uptime events.

    Better than Hetzner generally.

    @hacktek said:
    Netcup hosts in hetzner DCs

    No, they do not.

  • AC_FanAC_Fan Member

    @Zerpy said:

    @sonic said:
    How about Netcup uptime? Are their network solid?
    I couldnt find their status page to see uptime events.

    Better than Hetzner generally.

    @hacktek said:
    Netcup hosts in hetzner DCs

    No, they do not.

    The network is overall considered slightly better or equal to Hetzner's. And they have live migration for VMs, so uptime isn't that much of an issue.

    Yes, they do colocate in the Hetzner Nuremberg DCs.

  • ZerpyZerpy Member

    @AC_Fan said:
    Yes, they do colocate in the Hetzner Nuremberg DCs.

    hmmm, interesting - especially taking into account the latency from Netcup to Hetzner NBG seems high taken into account they're then within the same facility - they're only slightly less than the latency I have between two DCs being 50km apart

    Also considering Netcup is owned by Anexia, one would assume they'd use their parent company's infra

  • AC_FanAC_Fan Member

    @Zerpy said:

    @AC_Fan said:
    Yes, they do colocate in the Hetzner Nuremberg DCs.

    hmmm, interesting - especially taking into account the latency from Netcup to Hetzner NBG seems high taken into account they're then within the same facility - they're only slightly less than the latency I have between two DCs being 50km apart

    Also considering Netcup is owned by Anexia, one would assume they'd use their parent company's infra

    They only use Hetzner's colocation, they have their own network, so a stable latency of 2-3ms seems reasonable to me, since I believe it's over a private peering.
    NetCup seems like a sub-brand of Anexia that's kept fairly separate, since Anexia is very enterprise oriented.

  • sonicsonic Veteran

    @AC_Fan said:

    @Zerpy said:

    @sonic said:
    How about Netcup uptime? Are their network solid?
    I couldnt find their status page to see uptime events.

    Better than Hetzner generally.

    @hacktek said:
    Netcup hosts in hetzner DCs

    No, they do not.

    The network is overall considered slightly better or equal to Hetzner's. And they have live migration for VMs, so uptime isn't that much of an issue.

    Yes, they do colocate in the Hetzner Nuremberg DCs.

    Wow nice! Waiting for their BF offers to get some

  • nqservicesnqservices Member

    Thanks all for the advices and suggestions. Does anyone has a bech of Hetzner CCX31 or Netcup RS 8000 G9?

    Some said for me to get a dedicated server. The only thing with the Cloud is the probability of high uptime and not have to worry with hardware maintenance.

    I never had WordPress hosted on a dedicated server. Is the speed difference vs a cloud server high? Thanks

  • AC_FanAC_Fan Member
    edited November 2020

    @nqservices said:
    Thanks all for the advices and suggestions. Does anyone has a bech of Hetzner CCX31 or Netcup RS 8000 G9?

    Some said for me to get a dedicated server. The only thing with the Cloud is the probability of high uptime and not have to worry with hardware maintenance.

    I never had WordPress hosted on a dedicated server. Is the speed difference vs a cloud server high? Thanks

    Here you go:

    NetCup
    Hetzner

    Note: GeekBench 5 is roughly 900 and 4000 for Hetzner (single and multi thread, respectively) and 1000 and 6000 for NetCup.

  • Matix8981Matix8981 Member

    Why you dont won't choose dedicated server example AX41-NVME from hetzner, it's providing better performance and high clock processor with cheaper price.
    Both providers give a great services but netcup have a periodic contracts with penalties, but when you that want select from this offer, i will select hetzner because they have a hourly billing with scalable servers.

  • ZerpyZerpy Member

    @AC_Fan said:
    Note: GeekBench 5 is roughly 900 and 4000 for Hetzner (single and multi thread, respectively) and 1000 and 6000 for NetCup.

    8000 G9 shows 1042 (single) and 9517 (multi) ( https://browser.geekbench.com/v5/cpu/2316044 )

    Which is not bad at all, since that's what it would be as a physical server as well (the single core at least).

    For the sake of it, here's an up to date from hetzner CCX31:

    Sat Nov  7 17:14:36 CET 2020

Basic System Information:
---------------------------------
Processor  : Intel Xeon Processor (Skylake, IBRS)
CPU cores  : 8 @ 2095.078 MHz
AES-NI     : ✔ Enabled
VM-x/AMD-V : ❌ Disabled
RAM        : 30Gi
Swap       : 0B
Disk       : 226G

fio Disk Speed Tests (Mixed R/W 50/50):
---------------------------------
Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ----
Read       | 100.44 MB/s  (25.1k) | 1.15 GB/s    (18.1k)
Write      | 100.70 MB/s  (25.1k) | 1.16 GB/s    (18.2k)
Total      | 201.14 MB/s  (50.2k) | 2.32 GB/s    (36.3k)
           |                      |
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ----
Read       | 1.27 GB/s     (2.4k) | 1.27 GB/s     (1.2k)
Write      | 1.34 GB/s     (2.6k) | 1.35 GB/s     (1.3k)
Total      | 2.61 GB/s     (5.1k) | 2.63 GB/s     (2.5k)

iperf3 Network Speed Tests (IPv6):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed
                |                           |                 |
Clouvider       | London, UK (10G)          | 5.59 Gbits/sec  | 5.59 Gbits/sec
Online.net      | Paris, FR (10G)           | 7.74 Gbits/sec  | 5.69 Gbits/sec
WorldStream     | The Netherlands (10G)     | 11.3 Gbits/sec  | 7.07 Gbits/sec
Wifx            | Zurich, CH (10G)          | busy            | busy
Clouvider       | NYC, NY, US (10G)         | 1.64 Gbits/sec  | 1.56 Gbits/sec
Clouvider       | Los Angeles, CA, US (10G) | 1.14 Gbits/sec  | 1.09 Gbits/sec

Geekbench 5 Benchmark Test:
---------------------------------
Test            | Value
                |
Single Core     | 894
Multi Core      | 3647
Full Test       | https://browser.geekbench.com/v5/cpu/4585238
  • AC_FanAC_Fan Member
    edited November 2020

    @Zerpy That was on an empty node. If you take the last three benchmarks, the average is 894 and 6127.
    Source

  • umiumi Member

    If I need CPU power I'd go with Contabo. For more bandwidth Hetzner.

Sign In or Register to comment.