DigitalOcean, Vultr and Hetzner - Benchmarked each plan!

cyberduckcyberduck Provider
edited July 2020 in Technical

Hi guys,

This is my first post!

I have benchmarked most VPS plans on DigitalOcean, Vultr and Hetzner.

Here are my results:

DigitalOcean:

DigitalOcean $5: https://bin.privacytools.io/?23ca20216640cea7#yErXC6DoxvdoRAkLghDZ8m/Pbd8GtJ5sDIrFZzSxA6Q=

DigitalOcean $10: https://bin.privacytools.io/?3069db77ed5b5be9#7a0UG13s/ohBQcyZFOexuNOXSLDFmxLfvnC67XicQYU=

DigitalOcean $15: https://bin.privacytools.io/?bd60530c4b6332e2#HXYa+ns9h+vUrR/3UZ5Z6CooLEbjzqGR68k4L5hhNOg=

DigitalOcean $20: https://bin.privacytools.io/?942cb51ef2e7b423#fi2ty3ZEHGtYfS0FsK2adNHT1VWnGwsX+iXI6JBGWcQ=

DigitalOcean $40: https://bin.privacytools.io/?ee6d5fc601c4acfa#coMxzv1NH7bUMi/Rx+tkcRkNYbfDSKhS4IUMo2kG/No=

Vultr:

Vultr $5: https://bin.privacytools.io/?1f819cc75c7474ce#vLg8HWqUkkYSgNAB4jOr0bfNp0EJz3pr4UGjFBJwalY=

Vultr $10: https://bin.privacytools.io/?5547fb4c2a80101b#gX+QUIwagRGthUjn+aqf6ZGceGNeVNSUeFT0aE3oSDc=

Vultr $20: https://bin.privacytools.io/?b2e941f2a2d73646#hJSGgQ0UYOBJ/GdsvfMfV56P5shiG8E2u8J++0EB7qw=

Vultr $40: https://bin.privacytools.io/?37fd0bb3430afc50#KZk3wZk4kBHEDi/DKi6DEg2WKLxiotXsmxbsWJgjvj4=

Vultr $640: https://bin.privacytools.io/?246c068207398cc8#McGl3Uugm3GcoQv8yhJKQmw7NlnsvWbXjE4pV9VOo7c=

Vultr High Frequency

Vultr HF $6: https://bin.privacytools.io/?0104511948b3a2e3#JF595F8Js5d1n07wy0qRQCASR2E5xz0IuUawUCiSFXw=

Vultr HF $12: https://bin.privacytools.io/?e0f243240924ba25#lSHk0TPs8NbrcVyQKrOc+KtO72E6cVuqYAzLrUKjyRQ=

Vultr HF $24: https://bin.privacytools.io/?456d9b148023590d#VmThuIJwDTCiZdanS281+cweeeo6y519+BAwAMPznP8=

Hetzner:

Hetzner €2.99: https://bin.privacytools.io/?d191926282033e38#1KurHSEWy8bordyGy6AB41piwcIj3dZWJ+VadoB69dc=

Hetzner €4.19: https://bin.privacytools.io/?4d5378d726c930a9#rzD2oVk3vdR+iKetP9dvX3w9XF4ZWmAZUBC3ZbzPeEM=

Hetzner €8.28: https://bin.privacytools.io/?44c7eb2bf07fa02d#aTz9IfzcLaV1caRbICTRS+v4XwNzAPALHMqy3l4RIHk=

Hetzner €19.28: https://bin.privacytools.io/?dc9afd829fc6619d#02ZFguc7ucK9gvgf1U+EmeGBsYxanShz5nazaCXOgRE=

Hetzner €35.88: https://bin.privacytools.io/?eb52d24fb0cdca25#ORHUTFiiUYXDRr1zBjdNX6u0vZQHvLY4v8iu29ozguU=

Hostworld.uk

Hostworld £9.99: https://bin.privacytools.io/?8ff0c40abfc7ba4e#2t9ZkgCnzUqBcgOE9HlhqqKYSChA+DC+jHTSuwF08QA=

Benchmark comand used:

curl -s https://raw.githubusercontent.com/masonr/yet-another-bench-script/master/yabs.sh | bash

Hope you find this interesting and useful.

Cyberduck



edit:
12/07/2020 - Vultr High Frequency plans have been added!
14/07/2020 - Hostworld.uk VPS-3 added

Tagged:

Comments

  • seriesnseriesn OG

    Is this @cybertech or @cyberduck now I am truly confused!!

    Nexus Bytes Ryzen Powered NVMe VPS | NYC|Miami|LA|London|Netherlands| Singapore|Tokyo
    Storage VPS | LiteSpeed Powered Web Hosting + SSH access | Switcher Special |

  • cyberduckcyberduck Provider

    @seriesn said:
    Is this @cybertech or @cyberduck now I am truly confused!!

    Just a new dude on LES!

    Not related to @cybertech :)

  • sureiamsureiam

    Neat. Would be great if put into excel and then charted ;)

  • seriesnseriesn OG

    @cyberduck said:

    @seriesn said:
    Is this @cybertech or @cyberduck now I am truly confused!!

    Just a new dude on LES!

    Not related to @cybertech :)

    DAMN! Welcome to the gang yo! Now I can no longer call him Cyberduck .Sad day.

    Nexus Bytes Ryzen Powered NVMe VPS | NYC|Miami|LA|London|Netherlands| Singapore|Tokyo
    Storage VPS | LiteSpeed Powered Web Hosting + SSH access | Switcher Special |

  • MaxKVMMaxKVM
    edited July 2020

    This is seriously impressive if this is a full disk write (no caching).. which might be possible with Hetzner.

    @cyberduck said: Hetzner €35.88: https://bin.privacytools.io/?eb52d24fb0cdca25#ORHUTFiiUYXDRr1zBjdNX6u0vZQHvLY4v8iu29ozguU=

    Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 3.81 GB/s     (7.4k) | 4.04 GB/s     (3.9k)
Write      | 4.02 GB/s     (7.8k) | 4.31 GB/s     (4.2k)
Total      | 7.83 GB/s    (15.3k) | 8.35 GB/s     (8.1k)

    AMD EPYC powered Performance NVMe VPS - Los Angeles, Dallas, New York, Amsterdam, Singapore | Support | Status

  • evnixevnix OG

    @cyberduck said: Hope you find this interesting and useful.

    Vultr with all its marketing about high frequency CPUs seems to perform worse than Hetzner.

    My Personal Blog | Currently Building LoadMyCode

  • cyberduckcyberduck Provider

    @evnix said:

    @cyberduck said: Hope you find this interesting and useful.

    Vultr with all its marketing about high frequency CPUs seems to perform worse than Hetzner.

    Must be marketing bollocks!

    I personally run a VPS with Vultr, the host node died for 5 mins, twice within 48 hours!

    I also have a Hetzner VPS, with 100% uptime (created 2 months ago). It's a shame they only operate in the Germany and Finland.

  • poppzpoppz

    Thank you for this! Was recently looking to perform this myself but no longer necessary \o/

    Won't beat @seriesn amazingly performant VPS' though, for even less money :)

  • cybertechcybertech OGBenchmark King YABS 24/7/365

    HAHAHAHAHAHA

    I bench YABS 24/7/365 unless it's a leap year.

  • corbpiecorbpie OG

    @evnix said:

    @cyberduck said: Hope you find this interesting and useful.

    Vultr with all its marketing about high frequency CPUs seems to perform worse than Hetzner.

    OP did the old Vultr plans. Their new (always low stock now) High frequency plans spank the old shit.

    Not exactly a apples to apples test OP

  • seriesnseriesn OG

    @cybertech said:
    HAHAHAHAHAHA

    You find it funny?

    Nexus Bytes Ryzen Powered NVMe VPS | NYC|Miami|LA|London|Netherlands| Singapore|Tokyo
    Storage VPS | LiteSpeed Powered Web Hosting + SSH access | Switcher Special |

  • cybertechcybertech OGBenchmark King YABS 24/7/365

    @seriesn said:

    @cybertech said:
    HAHAHAHAHAHA

    You find it funny?

    No sir no. Not the benchmarks for sure. They were nicely done :tongue:

    I bench YABS 24/7/365 unless it's a leap year.

  • williewillie OG

    Thanks! Yes the vultr high frequency cpus are much faster than the regular ones. I'd also be interested in seeing the Hetzner EPYC plans.

  • DazzleDazzle OG

    Great. Thanks for your effort. Interested to see Upcloud too on the list.

    UpCloud free $25 through this aff link - Akamai, DigitalOcean and Vultr alternative, multiple location, IPv6.

  • vimalwarevimalware OG

    @willie said:
    I'd also be interested in seeing the Hetzner EPYC plans.

    They are listed. Check the 2nd one in hetzner list(4.19 eur incl VAT)

    Impressive nvme on all hetzner

  • cyberduckcyberduck Provider
    edited July 2020

    I have now added the Vultr High Frequency plans for those interested.

  • williewillie OG
    edited July 2020

    Oh thanks, I looked only at the smallest and biggest Hetzner (both Intel) and didn't notice that the 2nd and 3rd were Epyc. There is quite a discrepancy in speed between the 2nd and 3rd. I wonder what is up with that. I'd be interested in seeing the biggest (16 vcore) Epyc if the expense is tolerable.

    Having used both the regular and HF vultr instances I can say the HF are really snappy, much more than the 1.5x cpu benchmark difference makes it sound. They have very fast ssd's and at the time I tried them, the cpus were unloaded, at least on the small instances, so I was able to use a full core for a fairly long (maybe 1/2 hour) computation. Of course I'm sure a nonstop computation would have eventually gotten throttled.

  • DisputDisput

    OP @cyberduck how about BENCH MONSTER?> @corbpie said:

    @evnix said:

    @cyberduck said: Hope you find this interesting and useful.

    Vultr with all its marketing about high frequency CPUs seems to perform worse than Hetzner.

    OP did the old Vultr plans. Their new (always low stock now) High frequency plans spank the old shit.

    Not exactly a apples to apples test OP

    Could you please share some benches like yabs and monster from a new vult plans please?

  • cyberduckcyberduck Provider

    @Disput

    I have since added the new High Frequency Vultr plans.

    Also just added a BIG MONSTER plan for your enjoyment!

  • havochavoc OG

    When is Hetzner starting in the US?

  • Hetzner_OLHetzner_OL ProviderOG

    @havoc I'm not sure if we are ready for something like that...I think if we came to the US, there might be some reactions like...

    -- Katie ;)

    We're Katie and Lea and we'll do our best to answer questions you have about Hetzner Online. We and not our employer are responsible for any horrible puns and dated cultural references.

  • OujiOuji OG

    @Hetzner_OL said:
    @havoc I'm not sure if we are ready for something like that...I think if we came to the US, there might be some reactions like...

    -- Katie ;)

    That's exactly how we would react, Katie!

  • Hetzner_OLHetzner_OL ProviderOG
    edited July 2020

    Sorry. I have a zombie brain today apprently and accidentally posted the wrong thing here because I clicked on the wrong browser tab. :# --Katie

    We're Katie and Lea and we'll do our best to answer questions you have about Hetzner Online. We and not our employer are responsible for any horrible puns and dated cultural references.

  • seriesnseriesn OG

    @Hetzner_OL said:
    Sorry. I have a zombie brain today apprently and accidentally posted the wrong thing here because I clicked on the wrong browser tab. :# --Katie

    Shame on you!!

    Nexus Bytes Ryzen Powered NVMe VPS | NYC|Miami|LA|London|Netherlands| Singapore|Tokyo
    Storage VPS | LiteSpeed Powered Web Hosting + SSH access | Switcher Special |

  • moonswampmoonswamp
    > @MaxKVM said: Block Size 512k (IOPS) 1m (IOPS)
    Read 3.81 GB/s (7.4k) 4.04 GB/s (3.9k)
    Write 4.02 GB/s (7.8k) 4.31 GB/s (4.2k)
    Total 7.83 GB/s (15.3k) 8.35 GB/s (8.1k)

    that's super fast NVMe disk

    Waikey

  • williewillie OG

    I don't think you can add the read and write iops/sec together and report the total as meaning anything. Averaging them together makes more sense. 7000+/s is still very fast for a cheap vps. I thought it was more usual to measure iops at 512 bytes rather than 512k though.

  • MasonMason OG
    edited July 2020

    @willie said:
    I don't think you can add the read and write iops/sec together and report the total as meaning anything. Averaging them together makes more sense. 7000+/s is still very fast for a cheap vps. I thought it was more usual to measure iops at 512 bytes rather than 512k though.

    The read and write tests are being done concurrently, not independently. Both are added together to get the full IOPS load the disk was under during the test.

    Also, this test isn't really designed to show the absolutely maximum performance, but rather performance under somewhat real-world conditions.

    yabs daddy

  • williewillie OG

    Hmm, I'd want to see how that test works... maybe it is influenced somehow by caching. Otherwise, 16k/s = wow ;).

  • MasonMason OG

    @willie said:
    Hmm, I'd want to see how that test works... maybe it is influenced somehow by caching. Otherwise, 16k/s = wow ;).

    Shouldn't be affected by caching. Random read/writes and direct I/O flag on.

    It uses fio. Check it out, it's pretty neat stuff

    yabs daddy

  • vimalwarevimalware OG

    That reminds me. I need to go check up on my ultravps instance.
    I'm curious to see how much of that 2 week old 4GB test file I created, is still present in the Host Node's ZFS ARC(ram).

    5GB/s reads are a dead giveaway(even on nvme storage.)

  • vimalwarevimalware OG

    @vimalware said:
    That reminds me. I need to go check up on my ultravps instance.
    I'm curious to see how much of that 2 week old 4GB test file I created, is still present in the Host Node's ZFS ARC(ram).

    Here we go, for the curious. Looks like everything is travelling the full ZFS access path/stack.
    1 x Full Random read test with 128k blocksize :

    ./fio --randrepeat=1 --ioengine=libaio --direct=1 --gtod_reduce=1 --name=test --filename=test --bs=128k --iodepth=64 --size=4G --readwrite=randrw --rwmixread=100
test: (g=0): rw=randrw, bs=(R) 128KiB-128KiB, (W) 128KiB-128KiB, (T) 128KiB-128KiB, ioengine=libaio, iodepth=64
fio-3.20
Starting 1 process
Jobs: 1 (f=1): [r(1)][100.0%][r=445MiB/s][r=3563 IOPS][eta 00m:00s]
test: (groupid=0, jobs=1): err= 0: pid=27879: Fri Jul 24 02:16:33 2020
  read: IOPS=3957, BW=495MiB/s (519MB/s)(4096MiB/8281msec)
   bw (  KiB/s): min=257277, max=744750, per=100.00%, avg=508770.19, stdev=124657.26, samples=16
   iops        : min= 2009, max= 5818, avg=3974.31, stdev=973.96, samples=16
  cpu          : usr=13.99%, sys=74.13%, ctx=1187, majf=0, minf=6
  IO depths    : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=0.1%, 32=0.1%, >=64=99.8%
     submit    : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
     complete  : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.1%, >=64=0.0%
     issued rwts: total=32768,0,0,0 short=0,0,0,0 dropped=0,0,0,0
     latency   : target=0, window=0, percentile=100.00%, depth=64

Run status group 0 (all jobs):
   READ: bw=495MiB/s (519MB/s), 495MiB/s-495MiB/s (519MB/s-519MB/s), io=4096MiB (4295MB), run=8281-8281msec

Disk stats (read/write):
  vdb: ios=32399/2, merge=16/2, ticks=58520/0, in_queue=55212, util=84.67%

    I used 128k blocksize to bet that ultravps are using the default ZFS recordsize. I might be proven wrong in next test after a week.
    Summary: read: IOPS=3957, BW=495MiB/s (519MB/s)(4096MiB/8281msec)

    dstat output for the duration of the full Random read test. (to get some better visibility into CPU iowait under zfs.)

    --total-cpu-usage-- -dsk/total- -net/total- ---paging-- ---system--
usr sys idl wai stl| read  writ| recv  send|  in   out | int   csw 
--snipped--
  2   3  95   0   0|   0     0 |2039B  271B|   0     0 |  88   299 
  1   2  97   0   0|   0     0 |1611B  173B|   0     0 |  60   220 
 11  37  51   1   0|1672k  108k|3036B  338B|   0   108k| 229   541 
 23  67  10   0   0|  18M  628k|2511B  650B|   0   628k| 408   645 
 15  85   0   0   0| 657M    0 |2130B  720B|   0     0 |2585   572 
 17  82   1   0   0| 597M    0 |2512B  223B|   0     0 |2108   516 
 20  80   0   0   0| 552M    0 |2509B  396B|   0     0 |2318   554 
 18  82   0   0   0| 532M   32k|1807B  363B|   0     0 |2013   554 
 20  80   0   0   0| 392M  136k|2526B  372B|   0     0 |1435   582 
 19  81   0   0   0| 454M    0 |2335B  458B|   0     0 |1655   567 
 22  78   0   0   0| 344M    0 |2214B  383B|   0     0 |1303   646 
 22  78   0   0   0| 447M    0 |1769B  563B|   0     0 |1588   532 
 10  47  43   0   0| 103M    0 |2680B 1135B|   0     0 | 508   701 
  3   4  93   0   0|   0     0 |2333B  432B|   0     0 |  87   270 
  6   4  90   0   0|   0     0 |2346B 2078B|   0     0 |  94   332
--snipped--
  • williewillie OG

    Shouldn't be affected by caching. Random read/writes and direct I/O flag on.

    I have to wonder whether that flag really works under virtualization. It's sometimes doubtful even on bare metal, because of the many layers of caches in the controller, the drive, etc.

  • ViridWebViridWeb

    Wow! Hetzner's €2.99 plan is hard to defeat specially that performance at that price.

    Nice effort BTW,

    ViridWeb.com | /48 IPv6 rent or lease Starts from only $5 Yearly
    Onevirt.com - Netherlands VPS up to 10Gbps Epyc CPU starts from $5 Monthly

  • williewillie OG
    edited July 2020

    outdated, please delete.

  • williewillie OG
    edited July 2020

    I just ran the test on the 16 vcore Epyc instance. Multicore cpu was 5694 which I guess is disappointing since it's less than 10x the single core speed. It still beats the 8 vcore Intel instance by a fair amount. It's less than half the speed of a Hetzner Ryzen 3900 (AX61 I guess, 3.1 ghz, slower than the 3900X) dedi which scored 11656. So I guess it's slower even than a dedicated AX41. Oh well I guess it's still nice to be able to spin something like this up for a quick compute task, using the hourly billing. Other Hetzner results are all over the place:

    https://browser.geekbench.com/v5/cpu/search?utf8=✓&q=hetzner

    The AX51 (Ryzen 3700X) is consistently in the 9000+ range and the AX61 is 11000+. I haven't found an AX41 result that makes any sense yet but I'd guess it around 6000-7000.

    My bench results for the CPX51 are here:

    https://bin.privacytools.io/?32874f2c0832178f#w2UUx4hxYt/wEjltIRwc9eMUYbXtNW+KHjWBJyGy9rw=

    @cyberduck , please feel free to put this link in the main post.

    Hmm, here someone got 7449 on a CPX51 which is a fair bit better than what I got:

    https://browser.geekbench.com/v5/cpu/3006334

    Here, someone got 990 single, 12995 multi, so clearly an unloaded node. Better than 2x my instance's speed.

    https://browser.geekbench.com/v5/cpu/2787981

    So I guess these things are fairly loaded by now. My instance was at Falkenstein which might be the busiest site, fwiw.

  • vimalwarevimalware OG

    @willie said:

    my instance was at Falkenstein which might be the busiest site, fwiw.

    Yes I'd imagine Finland would be the place to try the one off 16cpu compute jobs

  • williewillie OG

    Yeah I might try Finland sometime but these things are too inconsistent and not that fast in the first place. I also wonder what happens in terms of throttling if you do a long-running compute task. The product is less interesting than I'd hoped. I still might use one now and then. Meanwhile my now ancient i7-3770 auction server is fast enough for most things (around 3500 on that bench).

    The €4.19 instance (3.49 ex vat) looks great as a general purpose vps fwiw.

Sign In or Register to comment.