Howdy, Stranger!

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


Avoro.eu changed VPS portfolio - Page 4
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.

Avoro.eu changed VPS portfolio

124»

Comments

  • jsgjsg Member, Resident Benchmarker

    @siemens said:

    @jsg said:

    @ascicode said:
    which of this 3 offers would you recommend? Or wait for php-friends deals?

    Hard to sensibly answer as those, pardon me, idiots only specified the Ryzen model. For the XEON and the EPYC model however only "XEON" and "EPYC" is specified.

    No clear model spec - no deal, thanks no.

    CPU models (except the Xeon) are listed on the same page the price is, not that hard to find.

    An offer on LET serves also and in particular the purpose to get LETians interested and to look at their website in the first place. "There you get the info on what we actually try to sell here" isn't the smartest way to go, in fact some like e.g. myself take that to mean "we don't care about and we don't respect you!" and stay away from that hoster.

    So, I'm pleased that @PHP_Friends finally to provide that information (and thanked them for it).

  • Thought I would share a yabs from my Avoro Rootserver M (6 cores Epyc 7513).

    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    #              Yet-Another-Bench-Script              #
    #                     v2024-01-01                    #
    # https://github.com/masonr/yet-another-bench-script #
    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
    
    Tue Feb 20 08:43:10 PM CET 2024
    
    Basic System Information:
    ---------------------------------
    Uptime     : 0 days, 0 hours, 11 minutes
    Processor  : AMD EPYC 7513 32-Core Processor
    CPU cores  : 6 @ 2599.998 MHz
    AES-NI     : ✔ Enabled
    VM-x/AMD-V : ✔ Enabled
    RAM        : 15.6 GiB
    Swap       : 1024.0 MiB
    Disk       : 314.9 GiB
    Distro     : Debian GNU/Linux 12 (bookworm)
    Kernel     : 6.1.0-17-amd64
    VM Type    : KVM
    IPv4/IPv6  : ✔ Online / ❌ Offline
    
    IPv4 Network Information:
    ---------------------------------
    ISP        : dataforest GmbH
    ASN        : AS58212 dataforest GmbH
    Host       : dataforest GmbH
    Location   : Offenbach Am Main, Hesse (HE)
    Country    : Germany
    
    fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/sda1):
    ---------------------------------
    Block Size | 4k            (IOPS) | 64k           (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 172.20 MB/s  (43.0k) | 1.73 GB/s    (27.1k)
    Write      | 172.66 MB/s  (43.1k) | 1.74 GB/s    (27.2k)
    Total      | 344.86 MB/s  (86.2k) | 3.48 GB/s    (54.3k)
               |                      |                     
    Block Size | 512k          (IOPS) | 1m            (IOPS)
      ------   | ---            ----  | ----           ---- 
    Read       | 6.99 GB/s    (13.6k) | 11.53 GB/s   (11.2k)
    Write      | 7.36 GB/s    (14.3k) | 12.29 GB/s   (12.0k)
    Total      | 14.36 GB/s   (28.0k) | 23.83 GB/s   (23.2k)
    
    iperf3 Network Speed Tests (IPv4):
    ---------------------------------
    Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping           
    -----           | -----                     | ----            | ----            | ----           
    Clouvider       | London, UK (10G)          | 6.87 Gbits/sec  | 1.47 Gbits/sec  | 11.1 ms        
    Scaleway        | Paris, FR (10G)           | 7.88 Gbits/sec  | 1.83 Gbits/sec  | 12.5 ms        
    NovoServe       | North Holland, NL (40G)   | 9.17 Gbits/sec  | 1.70 Gbits/sec  | 7.52 ms        
    Uztelecom       | Tashkent, UZ (10G)        | 1.76 Gbits/sec  | 313 Mbits/sec   | 99.8 ms        
    Clouvider       | NYC, NY, US (10G)         | 1.01 Gbits/sec  | 366 Mbits/sec   | 83.3 ms        
    Clouvider       | Dallas, TX, US (10G)      | 700 Mbits/sec   | 183 Mbits/sec   | 116 ms         
    Clouvider       | Los Angeles, CA, US (10G) | 561 Mbits/sec   | 264 Mbits/sec   | 141 ms         
    
    Geekbench 6 Benchmark Test:
    ---------------------------------
    Test            | Value                         
                    |                               
    Single Core     | 1091                          
    Multi Core      | 5082                          
    Full Test       | https://browser.geekbench.com/v6/cpu/4999408
    
    YABS completed in 12 min 8 sec
    
  • @chris32 said:

    fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/sda1):
    ---------------------------------
    Block Size | 4k            (IOPS) | 64k           (IOPS)
     ------   | ---            ----  | ----           ---- 
    Read       | 172.20 MB/s  (43.0k) | 1.73 GB/s    (27.1k)
    Write      | 172.66 MB/s  (43.1k) | 1.74 GB/s    (27.2k)
    Total      | 344.86 MB/s  (86.2k) | 3.48 GB/s    (54.3k)
              |                      |                     
    Block Size | 512k          (IOPS) | 1m            (IOPS)
     ------   | ---            ----  | ----           ---- 
    Read       | 6.99 GB/s    (13.6k) | 11.53 GB/s   (11.2k)
    Write      | 7.36 GB/s    (14.3k) | 12.29 GB/s   (12.0k)
    Total      | 14.36 GB/s   (28.0k) | 23.83 GB/s   (23.2k)
    

    That is very good! @PHP_Friends: you said that you use CEPH (https://lowendtalk.com/discussion/comment/3876246/#Comment_3876246). I thought CEPH was over network, which you said is 20 Gbps. How can it have over 144 Gbps write speed? Am I overlooking something?

    Single Core     | 1091
    Multi Core      | 5082
    

    That's also very good, especially the multi core performance.

  • Hmm, cheap plans

  • whats the actual model of the epyc rootservers? I'd like to check its info on google @PHP_Friends

  • A page ago its written.

  • @lukast__ said: 144 Gbps write speed

    * 190 Gbps rw speed (don't know how I got the 144, is wrong)

    To anyone who used/knows CEPH: is it possible to get higher read/write speeds than the physical network connection, perhaps over a cache?

  • dataforestdataforest Member, Patron Provider

    @lukast__ said:

    @chris32 said:

    fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/sda1):
    ---------------------------------
    Block Size | 4k            (IOPS) | 64k           (IOPS)
     ------   | ---            ----  | ----           ---- 
    Read       | 172.20 MB/s  (43.0k) | 1.73 GB/s    (27.1k)
    Write      | 172.66 MB/s  (43.1k) | 1.74 GB/s    (27.2k)
    Total      | 344.86 MB/s  (86.2k) | 3.48 GB/s    (54.3k)
              |                      |                     
    Block Size | 512k          (IOPS) | 1m            (IOPS)
     ------   | ---            ----  | ----           ---- 
    Read       | 6.99 GB/s    (13.6k) | 11.53 GB/s   (11.2k)
    Write      | 7.36 GB/s    (14.3k) | 12.29 GB/s   (12.0k)
    Total      | 14.36 GB/s   (28.0k) | 23.83 GB/s   (23.2k)
    

    That is very good! @PHP_Friends: you said that you use CEPH (https://lowendtalk.com/discussion/comment/3876246/#Comment_3876246). I thought CEPH was over network, which you said is 20 Gbps. How can it have over 144 Gbps write speed? Am I overlooking something?

    Single Core     | 1091
    Multi Core      | 5082
    

    That's also very good, especially the multi core performance.

    20 GBit/s is the internet connection, the Ceph and Internet are of course strictly separated.

    Thanked by 1lukast__
  • @PHP_Friends said: 20 GBit/s is the internet connection, the Ceph and Internet are of course strictly separated.

    So you have 2 x 10 Gbit/s connection for internet, and seperate connection(s) for Ceph?
    And even if someone is writing/reading much, it doesn't affect the internet connection?
    That makes sense, thanks.
    (Sorry if that was a stupid question, I know almost nothing about Ceph.)

  • dataforestdataforest Member, Patron Provider

    @lukast__ said:

    @PHP_Friends said: 20 GBit/s is the internet connection, the Ceph and Internet are of course strictly separated.

    So you have 2 x 10 Gbit/s connection for internet, and seperate connection(s) for Ceph?
    And even if someone is writing/reading much, it doesn't affect the internet connection?
    That makes sense, thanks.
    (Sorry if that was a stupid question, I know almost nothing about Ceph.)

    Correct :) If you have any questions, please send a PN or ticket :)

    Thanked by 1lukast__
  • WhizzWrWhizzWr Member

    @Avoro @PHP_Friends are you also considering offering high capacity storage as alternative to fast NVME?

    Like 1TB, 2TB capacity, HDD or even mountable block storage is fine.

  • dataforestdataforest Member, Patron Provider

    @WhizzWr said:
    @Avoro @PHP_Friends are you also considering offering high capacity storage as alternative to fast NVME?

    Like 1TB, 2TB capacity, HDD or even mountable block storage is fine.

    Not right now :)

    Thanked by 1WhizzWr
  • idleidle Member

    @PHP_Friends
    What is the CPU fair use policy for Avoro VPS (limits, penalties, etc)?

  • marufmaruf Member

    @PHP_Friends said:

    @WhizzWr said:
    @Avoro @PHP_Friends are you also considering offering high capacity storage as alternative to fast NVME?

    Like 1TB, 2TB capacity, HDD or even mountable block storage is fine.

    Not right now :)

    what's the bandwidth limit for avoro vps ?

  • conceptconcept Member

    @maruf said:
    what's the bandwidth limit for avoro vps ?

    1Gbit Guranteed, 10gbit burst

    https://lowendtalk.com/discussion/comment/3461557/#Comment_3461557

    Thanked by 2maruf dataforest
  • dataforestdataforest Member, Patron Provider

    @idle said:
    @PHP_Friends
    What is the CPU fair use policy for Avoro VPS (limits, penalties, etc)?

    No penalties, but its not dedicated.

    Thanked by 1idle
  • lukast__lukast__ Member

    @PHP_Friends: on your "Technische Beschränkungen" (Technical Limitations) page you say that benchmarks are not allowed. Is this also the case with the rootservers, since the cores are dedicated there?
    And in your TOS you say that you don't own your servers (§3)? I would think that is outdated?
    And does §7.5 also apply to hosting the (private) website of a friend (if so, that's no problem, just want to know)?

  • @lukast__ said:
    @PHP_Friends: on your "Technische Beschränkungen" (Technical Limitations) page you say that benchmarks are not allowed. Is this also the case with the rootservers, since the cores are dedicated there?
    And in your TOS you say that you don't own your servers (§3)? I would think that is outdated?
    And does §7.5 also apply to hosting the (private) website of a friend (if so, that's no problem, just want to know)?

    oh! benchmarks = yabs not allowed @PHP_Friends ? please clarify!

    Thanked by 1COLBYLICIOUS
  • fendixfendix Member

    @hyperblast said: oh! benchmarks = yabs not allowed @PHP_Friends ? please clarify!

    I asked myself the same question a few years ago. As long as you have dedicated cores, you can basically do whatever you want.

    I received the following reply on Twitter in 2021:

    Shared cores:
    Basically there are no limits, but there are products with dedicated cores and those without - so you may not always have all cores available. In addition, there are automatic notifications in the event of unusual loads, but these are purely informative. And it takes more than a few minutes of full load to trigger them.

    Dedicated cores:
    There are no limits :) The product has dedicated cores.

    (Grundsätzlich gibt's keine Limits, aber es gibt Produkte mit dedizierten Kernen und solche ohne - da kann es halt sein, dass dir nicht immer alle Kerne zur Verfügung stehen. Darüber hinaus gibt es automatische Hinweise bei ungewöhnlicher Last, die jedoch rein informativ sind.)

    Thanked by 2t0m hyperblast
  • @fendix said:

    @hyperblast said: oh! benchmarks = yabs not allowed @PHP_Friends ? please clarify!

    I asked myself the same question a few years ago. As long as you have dedicated cores, you can basically do whatever you want.

    I received the following reply on Twitter in 2021:

    Shared cores:
    Basically there are no limits, but there are products with dedicated cores and those without - so you may not always have all cores available. In addition, there are automatic notifications in the event of unusual loads, but these are purely informative. And it takes more than a few minutes of full load to trigger them.

    Dedicated cores:
    There are no limits :) The product has dedicated cores.

    (Grundsätzlich gibt's keine Limits, aber es gibt Produkte mit dedizierten Kernen und solche ohne - da kann es halt sein, dass dir nicht immer alle Kerne zur Verfügung stehen. Darüber hinaus gibt es automatische Hinweise bei ungewöhnlicher Last, die jedoch rein informativ sind.)

    @PHP_Friends @Avoro what is your current position on this?

  • dataforestdataforest Member, Patron Provider

    @hyperblast said:

    @fendix said:

    @hyperblast said: oh! benchmarks = yabs not allowed @PHP_Friends ? please clarify!

    I asked myself the same question a few years ago. As long as you have dedicated cores, you can basically do whatever you want.

    I received the following reply on Twitter in 2021:

    Shared cores:
    Basically there are no limits, but there are products with dedicated cores and those without - so you may not always have all cores available. In addition, there are automatic notifications in the event of unusual loads, but these are purely informative. And it takes more than a few minutes of full load to trigger them.

    Dedicated cores:
    There are no limits :) The product has dedicated cores.

    (Grundsätzlich gibt's keine Limits, aber es gibt Produkte mit dedizierten Kernen und solche ohne - da kann es halt sein, dass dir nicht immer alle Kerne zur Verfügung stehen. Darüber hinaus gibt es automatische Hinweise bei ungewöhnlicher Last, die jedoch rein informativ sind.)

    @PHP_Friends @Avoro what is your current position on this?

    Benchmarks are of course allowed, but the point here is that it is not allowed to use benchmarks to overload the respective systems permanently. In the meantime, every one of our customers actually uses a benchmark to test performance and we even link to benchmark commands on our pages.

Sign In or Register to comment.