Howdy, Stranger!

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


Contabo new raw NVMe speed product line benchmark/review - Page 3
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.

Contabo new raw NVMe speed product line benchmark/review

13»

Comments

  • dev_vpsdev_vps Member
    edited September 2021

    @jsg said:

    Accordingly I will ask Contabo to remove my quote from their site.

    I've said it already but I will repeat it: My first loyalty is with the LET community and not with any provider. And my integrity is not for sale.

    Respect.

    I have cancelled my contabo VPS and asked for money back using 14 day rule.

    My experience with $36/year vps (inception hosting @Clouvider) is much better and I plan to renew the VPS for the next year.

  • jsgjsg Member, Resident Benchmarker

    @dev_vps said:
    My experience with $36/year vps (inception hosting @Clouvider) is much better and I plan to renew the VPS for the next year.

    Yes, IH has some very tasty products. I still like my small promo VPS very much.

  • bruh21bruh21 Member, Host Rep

    it seems i would not bother wasting my time with the nvme
    the ssd line seems like it is still the superior deal if you want a contabo server

  • ArkasArkas Moderator
    edited September 2021

    @bruh21 said: it seems i would not bother wasting my time with the nvme
    the ssd line seems like it is still the superior deal if you want a contabo server

    I don't think they have new offers on SSD, it's only NVME for new orders.

  • bruh21bruh21 Member, Host Rep

    @Arkas said:

    @bruh21 said: it seems i would not bother wasting my time with the nvme
    the ssd line seems like it is still the superior deal if you want a contabo server

    I don't think they have new offers on SSD, it's only NVME for new orders.

    no, when you click order it still has the 200gb ssd option

    Thanked by 1Arkas
  • @AXYZE said:

    @mindanalyzer said:
    I moved from a contabo VPS L SSD to a VPS L NVMe (had to set it from scratch because Contabo's rescaling capabilities are non-existent) and the latter is ridiculously slow in terms of storage access. Even ran a comparison with a much lower specs hetzner VPS and the difference is night to day

    What's going on, what am I missing?

    In terms of software running, I setup ubuntu 20.04, hardened it using standard procedures and I am running some apps that also run with other providers. Like I said, I turned everything off and ran that comparison and got really surprised at how crappy my results were

    What am I missing? Have I been lied to?

    Youre missing nothing, normal case with Contabo.

    Stick with Hetzner/Contabo/PHP-Friends if you want quality on budget. Better performance with same money, no setup fees (so its even cheaper), working DDoS Protection etc.

    Buy Contabo only if you need SSD on cheap. NVMe plan is not cost effective, you will get more, faster storage with netcup.

    I should write "stick with Hetzner/netcup/PHP-Friends".
    Cant edit now, fuck.

    Thanked by 1TODO
  • ArkasArkas Moderator
    edited September 2021

    @AXYZE said: I should write "stick with Hetzner/netcup/PHP-Friends".
    Cant edit now, fuck.

    What about Leaseweb? I mean other than the pricing.

  • @Arkas said:

    @AXYZE said: I should write "stick with Hetzner/netcup/PHP-Friends".
    Cant edit now, fuck.

    What about Leaseweb? I mean other than the pricing.

    For dedis: Hetzner/Reliablesite/Leaseweb/OVH/SYS
    No idea about Leaseweb VPS offer.

    Thanked by 1Arkas
  • jsgjsg Member, Resident Benchmarker
    edited September 2021

    News:

    As I've said I did contact Contabo and asked them to either delete the quote from me or to delete the "three times". And they complied and took the "three times" out, so the current version just says "faster than" (instead of "three times faster than") which is correct albeit based on only one VM (my private one which I pay for).

    I've also confronted them with my current results as gathered from my own private VPS, which show numbers far below what my original benchmark showed, even on a production node.

    Now I've just got another communications in which some manager confirms that there indeed was a problem with some interface hardware. They've tracked down the problem and I'm benchmarking it while we talk.

    I also confronted them with the allegations of them brutally overselling and they assured me that those are not true and that they do not overcrowd their systems.

    Now, of course one shouldn't just believe what a company says about itself, but on the other hand I've never found them to be lying to me and in fact we have quite frank conversations and they have been surprisingly open about quite a few things I've asked.

    Does that mean that I again recommend Contabo? No. At least not yet. I want to dig a bit deeper and to benchmark the VPS with the fixed interface problem. As I said they seem to have never lied to me, but on the other hand I can't and won't just ignore multiple LET users who speak about Contabo brutally overselling.

    So, while things seem to move in a good direction, at least for now I go with those of you who prefer and recommend other providers like NexusBytes / @seriesn (not german but really great) or netcup both of which I have a VPS with and a good or even excellent customer experience or Hetzner / @Hetzner_OL with whom I have no personal experience but who have such a large and consistent fan club here that I can't but presume that they are a fine provider too.

    Thanked by 1Arkas
  • Just got email from contabo that it may take “several days” before the refund can be issued.

    Personally, Contabo is not even in my top 10 list.

    Most value/performance combo -
    1. Inception Hosting
    2. Racknerd

  • dev_vpsdev_vps Member
    edited September 2021

    @jsg said:
    News:

    As I've said I did contact Contabo and asked them to either delete the quote from me or to delete the "three times". And they complied and took the "three times" out, so the current version just says "faster than" (instead of "three times faster than") which is correct albeit based on only one VM (my private one which I pay for).

    The delete “three times” part was not disclosed before in the forum. You simply stated that you have asked them to remove the quote.

    You have yourself admitted that “faster than” is based on just one VPS. It seems you are comfortable putting your just one vps based quote on the site that acts a marketing magnet.

    Thanked by 2adly TODO
  • jsgjsg Member, Resident Benchmarker

    @dev_vps said:

    @jsg said:
    News:

    As I've said I did contact Contabo and asked them to either delete the quote from me or to delete the "three times". And they complied and took the "three times" out, so the current version just says "faster than" (instead of "three times faster than") which is correct albeit based on only one VM (my private one which I pay for).

    The delete “three times” part was not disclosed before in the forum. You simply stated that you have asked them to remove the quote.

    You have yourself admitted that “faster than” is based on just one VPS. It seems you are comfortable putting your just one vps based quote on the site that acts a marketing magnet.

    Well, provide me with 10 or 100 VPS and I'll test them. I happen to have just 1 and based on that VPS the quote (in its changed form) is OK. IF I should find them lying and/or cheating on me, having to pull the quote completely will be the smallest problem they'll have with me.

    Sorry, it just doesn't work the way some of you want it. I very much dislike a certain provider yet when he had a promo with a really good product I gave them a fair review. Because the emotions of a benchmarker must not guide his work and because integrity and fairness is of paramount importance. He may be an a__hole (IMO) but his VPS was nice and fast and a bargain and so that's what I reported, period.
    Same game in the other direction. I have publicly admitted to be a big NexusBytes / @seriesn fan, but when I review a product of his I present cold data.
    A benchmarker MUST be fair and report what (s)he sees, no matter whether he likes or dislikes a provider.

    Actually Contabo is a good example. On a VPS with the problem fixed I suddenly see fabulous numbers again - yet what did I do? Did I go back to recommending them? Nope. One (1) suddenly fast again VPS doesn't cut it anymore after what I experienced. That one fast again VPS might be nice for me personally but I do not accept that as a basis to recommend them again. And I told them so. I want to see evidence that clearly indicates that all NVMe VPS nodes are fixed and fast.

  • Ordered a VPS L NVMe with 400GB NVMe (location EU) for 17,84 € per month on a 12 month plan. The results are not satisfying and I will cancel the contract within the 14 days period.

    -------------------------------------------------
     nench.sh v2019.07.20 -- https://git.io/nench.sh
     benchmark timestamp:    2022-10-10 02:06:53 UTC
    -------------------------------------------------
    
    Processor:    AMD EPYC 7282 16-Core Processor
    CPU cores:    8
    Frequency:    2794.748 MHz
    RAM:          29Gi
    Swap:         4.0Gi
    Kernel:       Linux 5.4.0-126-generic x86_64
    
    Disks:
    sda    400G  SSD
    
    CPU: SHA256-hashing 500 MB
        4.547 seconds
    CPU: bzip2-compressing 500 MB
        9.450 seconds
    CPU: AES-encrypting 500 MB
        2.669 seconds
    
    ioping: seek rate
        min/avg/max/mdev = 32.4 us / 424.4 us / 36.2 ms / 1.38 ms
    ioping: sequential read speed
        generated 11.4 k requests in 5.01 s, 2.78 GiB, 2.27 k iops, 568.8 MiB/s
    
    dd: sequential write speed
        1st run:    322.34 MiB/s
        2nd run:    732.42 MiB/s
        3rd run:    807.76 MiB/s
        average:    620.84 MiB/s
    
    IPv4 speedtests
        your IPv4:    75.xxx.xxx.xxxx
    
        Cachefly CDN:         64.23 MiB/s
        Leaseweb (NL):        59.87 MiB/s
        Softlayer DAL (US):   5.44 MiB/s
        Online.net (FR):      63.12 MiB/s
        OVH BHS (CA):         10.00 MiB/s
    
    No IPv6 connectivity detected
    -------------------------------------------------
    

    In the meantime my Netcup RS 4000 G9 SE (800GB SSD) for 24,42€ per month is doing this:


     nench.sh v2019.07.20 -- https://git.io/nench.sh
     benchmark timestamp:    2022-10-10 02:06:54 UTC
    -------------------------------------------------
    
    Processor:    AMD EPYC 7702P 64-Core Processor
    CPU cores:    8
    Frequency:    1996.249 MHz
    RAM:          31Gi
    Swap:         4,0Gi
    Kernel:       Linux 5.4.0-126-generic x86_64
    
    Disks:
    sda    800G  SSD
    
    CPU: SHA256-hashing 500 MB
        2,635 seconds
    CPU: bzip2-compressing 500 MB
        5,458 seconds
    CPU: AES-encrypting 500 MB
        1,058 seconds
    
    ioping: seek rate
        min/avg/max/mdev = 41.8 us / 70.1 us / 16.2 ms / 174.0 us
    ioping: sequential read speed
        generated 28.1 k requests in 5.00 s, 6.86 GiB, 5.62 k iops, 1.37 GiB/s
    
    dd: sequential write speed
        1st run:    610.35 MiB/s
        2nd run:    705.72 MiB/s
        3rd run:    899.31 MiB/s
        average:    738.46 MiB/s
    
    IPv4 speedtests
        your IPv4:    89.xxx.xxx.xxxx
    
        Cachefly CDN:         195.69 MiB/s
        Leaseweb (NL):        76.05 MiB/s
        Softlayer DAL (US):   12.48 MiB/s
        Online.net (FR):      168.49 MiB/s
        OVH BHS (CA):         14.80 MiB/s
    
    No IPv6 connectivity detected
    -------------------------------------------------
    
  • AXYZEAXYZE Member
    edited October 2022

    @subven said:
    Ordered a VPS L NVMe with 400GB NVMe (location EU) for 17,84 € per month on a 12 month plan. The results are not satisfying and I will cancel the contract within the 14 days period.

    -------------------------------------------------
     nench.sh v2019.07.20 -- https://git.io/nench.sh
     benchmark timestamp:    2022-10-10 02:06:53 UTC
    -------------------------------------------------
    
    Processor:    AMD EPYC 7282 16-Core Processor
    CPU cores:    8
    Frequency:    2794.748 MHz
    RAM:          29Gi
    Swap:         4.0Gi
    Kernel:       Linux 5.4.0-126-generic x86_64
    
    Disks:
    sda    400G  SSD
    
    CPU: SHA256-hashing 500 MB
        4.547 seconds
    CPU: bzip2-compressing 500 MB
        9.450 seconds
    CPU: AES-encrypting 500 MB
        2.669 seconds
    
    ioping: seek rate
        min/avg/max/mdev = 32.4 us / 424.4 us / 36.2 ms / 1.38 ms
    ioping: sequential read speed
        generated 11.4 k requests in 5.01 s, 2.78 GiB, 2.27 k iops, 568.8 MiB/s
    
    dd: sequential write speed
        1st run:    322.34 MiB/s
        2nd run:    732.42 MiB/s
        3rd run:    807.76 MiB/s
        average:    620.84 MiB/s
    
    IPv4 speedtests
        your IPv4:    75.xxx.xxx.xxxx
    
        Cachefly CDN:         64.23 MiB/s
        Leaseweb (NL):        59.87 MiB/s
        Softlayer DAL (US):   5.44 MiB/s
        Online.net (FR):      63.12 MiB/s
        OVH BHS (CA):         10.00 MiB/s
    
    No IPv6 connectivity detected
    -------------------------------------------------
    

    In the meantime my Netcup RS 4000 G9 SE (800GB SSD) for 24,42€ per month is doing this:


     nench.sh v2019.07.20 -- https://git.io/nench.sh
     benchmark timestamp:    2022-10-10 02:06:54 UTC
    -------------------------------------------------
    
    Processor:    AMD EPYC 7702P 64-Core Processor
    CPU cores:    8
    Frequency:    1996.249 MHz
    RAM:          31Gi
    Swap:         4,0Gi
    Kernel:       Linux 5.4.0-126-generic x86_64
    
    Disks:
    sda    800G  SSD
    
    CPU: SHA256-hashing 500 MB
        2,635 seconds
    CPU: bzip2-compressing 500 MB
        5,458 seconds
    CPU: AES-encrypting 500 MB
        1,058 seconds
    
    ioping: seek rate
        min/avg/max/mdev = 41.8 us / 70.1 us / 16.2 ms / 174.0 us
    ioping: sequential read speed
        generated 28.1 k requests in 5.00 s, 6.86 GiB, 5.62 k iops, 1.37 GiB/s
    
    dd: sequential write speed
        1st run:    610.35 MiB/s
        2nd run:    705.72 MiB/s
        3rd run:    899.31 MiB/s
        average:    738.46 MiB/s
    
    IPv4 speedtests
        your IPv4:    89.xxx.xxx.xxxx
    
        Cachefly CDN:         195.69 MiB/s
        Leaseweb (NL):        76.05 MiB/s
        Softlayer DAL (US):   12.48 MiB/s
        Online.net (FR):      168.49 MiB/s
        OVH BHS (CA):         14.80 MiB/s
    
    No IPv6 connectivity detected
    -------------------------------------------------
    

    Lol CPU more than twice as fast (AES) on netcup while both are EPYCs 2nd gen, 8vCPU and netcup has A LOT lower frequency.

    We are overselling to the max,
    Contabo

    Thanked by 1TODO
  • I lol'd on complaints... Do you even utilize 100% that humongous nvme speed for decent period of time or just looking at bench without real world usage?

  • I left Contabo this month after 2 years.

Sign In or Register to comment.