Alternativa ke Contabo VPS

[Ymzk]

Taky nějak nerozumím tomu, proč ne Hetzner, ale když ne on, tak Netcup nebo PHP-Friends.

Pozeram ten Netcup a vyzera to slubne. Hlavne ten root server co oni nazyvaju. Taka lepsia "VPS".
Ma tam niekto nieco? Dokazal by tam niekto hodit fio a otestovat? Ked nie tak to vyskusam. Ale nechapem preco, túto sluzbu nepoznam.

[Reklama]

[mhepp]

Kvůli diverzitě mám i něco u Vultr.com - jsou to amíci, ale datacentra mají i v EU. Nepočítej ale se stejným výkonem jako Contabo. Je to slabší.

Tak to uz musi system bootovat tak hodinu...
Mal som u contabo VPS a ten disk performance bol strasny, aj login na ssh trval okolo 10s (prave nainstalovana fedora), chvilu som to pouzival na nextcloud, siet bola v pohode, uzkym hrdlom bol prave disk.

Presiel som od nich k VPSfree a som zatial spokojny

Nevím, kde co jak máš, ale odezva na login je téměř okamžitá, do sekundy určitě, reboot do cca minuty. Oba, jak Vultr, tak Contabo.

Ne každý potřebuje pro svůj stroj hvězdný diskový výkon, někdo chce raději více RAM a CPU.

Každopádně, dívám se, že konkurence k Contabo už je a je docela zajímavá - i když je trochu dražší... Asi to promyslím...

[Ymzk]

Kvůli diverzitě mám i něco u Vultr.com - jsou to amíci, ale datacentra mají i v EU. Nepočítej ale se stejným výkonem jako Contabo. Je to slabší.

Tak to uz musi system bootovat tak hodinu...
Mal som u contabo VPS a ten disk performance bol strasny, aj login na ssh trval okolo 10s (prave nainstalovana fedora), chvilu som to pouzival na nextcloud, siet bola v pohode, uzkym hrdlom bol prave disk.

Presiel som od nich k VPSfree a som zatial spokojny

Nevím, kde co jak máš, ale odezva na login je téměř okamžitá, do sekundy určitě, reboot do cca minuty. Oba, jak Vultr, tak Contabo.

Ne každý potřebuje pro svůj stroj hvězdný diskový výkon, někdo chce raději více RAM a CPU.

Každopádně, dívám se, že konkurence k Contabo už je a je docela zajímavá - i když je trochu dražší... Asi to promyslím...

Mozes mat aj 1TB RAM, aj CPU vykonu ale ked je disk pomalsi ako SD karta je ti to totalne k nicomu.

[zdfs]

Na fóru Netcup mají dokonce tabulku, kde uživatelé měří IOPS.

https://forum.netcup.de/administration-eines-server-vserver/vserver-server-kvm-server/12606-messen-von-io-performance-bzw-iops-performance-der-ssd/

https://docs.google.com/spreadsheets/d/1w38zM0Bwbd4VdDCQoi1buo2I-zpwg8e0wVzFGSPh3iE/edit#gid=0

Contabo Cloud VPS S NVMe (bez libaio parametru, kvůli Windows Server 2022)

Kód: [Vybrat]

fio --randrepeat=1 --direct=1 --gtod_reduce=1 --name=test1 --filename=test1 --bs=4k --iodepth=64 --size=4G --readwrite=randrw --rwmixread=75
fio: this platform does not support process shared mutexes, forcing use of threads. Use the 'thread' option to get rid of this warning.
test1: (g=0): rw=randrw, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=windowsaio, iodepth=64
fio-3.33
Starting 1 thread
test1: Laying out IO file (1 file / 4096MiB)
Jobs: 1 (f=1): [m(1)][100.0%][r=59.7MiB/s,w=19.5MiB/s][r=15.3k,w=5000 IOPS][eta 00m:00s]
test1: (groupid=0, jobs=1): err= 0: pid=9524: Sun Nov 20 08:02:15 2022
  read: IOPS=16.4k, BW=63.9MiB/s (67.0MB/s)(3070MiB/48043msec)
   bw (  KiB/s): min=10240, max=170541, per=100.00%, avg=65552.71, stdev=23472.02, samples=94
   iops        : min= 2560, max=42635, avg=16387.87, stdev=5868.01, samples=94
  write: IOPS=5467, BW=21.4MiB/s (22.4MB/s)(1026MiB/48043msec); 0 zone resets
   bw (  KiB/s): min= 3472, max=56857, per=100.00%, avg=21897.66, stdev=7796.48, samples=94
   iops        : min=  868, max=14214, avg=5474.15, stdev=1949.10, samples=94
  cpu          : usr=6.24%, sys=27.06%, ctx=0, majf=0, minf=0
  IO depths    : 1=0.1%, 2=0.1%, 4=0.2%, 8=1.0%, 16=3.2%, 32=52.6%, >=64=42.9%
     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=97.9%, 8=1.4%, 16=0.3%, 32=0.1%, 64=0.2%, >=64=0.0%
     issued rwts: total=785920,262656,0,0 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=63.9MiB/s (67.0MB/s), 63.9MiB/s-63.9MiB/s (67.0MB/s-67.0MB/s), io=3070MiB (3219MB), run=48043-48043msec
  WRITE: bw=21.4MiB/s (22.4MB/s), 21.4MiB/s-21.4MiB/s (22.4MB/s-22.4MB/s), io=1026MiB (1076MB), run=48043-48043msec

Kód: [Vybrat]

fio --randrepeat=1 --direct=1 --gtod_reduce=1 --name=test --filename=random_read_write.fio --bs=4k --iodepth=64 --size=4G --readwrite=randrw --rwmixread=80
fio: this platform does not support process shared mutexes, forcing use of threads. Use the 'thread' option to get rid of this warning.
test: (g=0): rw=randrw, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=windowsaio, iodepth=64
fio-3.33
Starting 1 thread
test: Laying out IO file (1 file / 4096MiB)
Jobs: 1 (f=1): [m(1)][100.0%][r=80.9MiB/s,w=19.5MiB/s][r=20.7k,w=5003 IOPS][eta 00m:00s]
test: (groupid=0, jobs=1): err= 0: pid=9452: Sun Nov 20 08:07:29 2022
  read: IOPS=22.2k, BW=86.6MiB/s (90.8MB/s)(3276MiB/37815msec)
   bw (  KiB/s): min= 4064, max=199206, per=100.00%, avg=88997.93, stdev=31191.93, samples=74
   iops        : min= 1016, max=49801, avg=22249.24, stdev=7797.99, samples=74
  write: IOPS=5554, BW=21.7MiB/s (22.7MB/s)(820MiB/37815msec); 0 zone resets
   bw (  KiB/s): min= 1168, max=49643, per=100.00%, avg=22285.32, stdev=7745.82, samples=74
   iops        : min=  292, max=12410, avg=5571.11, stdev=1936.46, samples=74
  cpu          : usr=5.29%, sys=29.09%, ctx=0, majf=0, minf=0
  IO depths    : 1=0.1%, 2=0.1%, 4=0.3%, 8=1.2%, 16=4.1%, 32=61.1%, >=64=33.2%
     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=97.0%, 8=1.9%, 16=0.6%, 32=0.3%, 64=0.3%, >=64=0.0%
     issued rwts: total=838551,210025,0,0 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=86.6MiB/s (90.8MB/s), 86.6MiB/s-86.6MiB/s (90.8MB/s-90.8MB/s), io=3276MiB (3435MB), run=37815-37815msec
  WRITE: bw=21.7MiB/s (22.7MB/s), 21.7MiB/s-21.7MiB/s (22.7MB/s-22.7MB/s), io=820MiB (860MB), run=37815-37815msec
OVH Starter Server za 12 € na 12 měsíců, Ubuntu 22.04

Kód: [Vybrat]

fio --randrepeat=1 --ioengine=libaio --direct=1 --gtod_reduce=1 --name=test1 --filename=test1 --bs=4k --iodepth=64 --size=4G --readwrite=randrw --rwmixread=75
test1: (g=0): rw=randrw, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=libaio, iodepth=64
fio-3.28
Starting 1 process
test1: Laying out IO file (1 file / 4096MiB)
Jobs: 1 (f=1): [m(1)][100.0%][r=23.5MiB/s,w=7859KiB/s][r=6004,w=1964 IOPS][eta 00m:00s]
test1: (groupid=0, jobs=1): err= 0: pid=54221: Sun Nov 20 08:05:31 2022
  read: IOPS=6003, BW=23.5MiB/s (24.6MB/s)(3070MiB/130907msec)
   bw (  KiB/s): min=23752, max=28792, per=100.00%, avg=24034.02, stdev=297.67, samples=261
   iops        : min= 5938, max= 7198, avg=6008.51, stdev=74.42, samples=261
  write: IOPS=2006, BW=8026KiB/s (8218kB/s)(1026MiB/130907msec); 0 zone resets
   bw (  KiB/s): min= 7120, max=10072, per=100.00%, avg=8032.98, stdev=333.48, samples=261
   iops        : min= 1780, max= 2518, avg=2008.25, stdev=83.37, samples=261
  cpu          : usr=3.49%, sys=13.39%, ctx=680085, majf=0, minf=8
  IO depths    : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=0.1%, 32=0.1%, >=64=100.0%
     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=785920,262656,0,0 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=23.5MiB/s (24.6MB/s), 23.5MiB/s-23.5MiB/s (24.6MB/s-24.6MB/s), io=3070MiB (3219MB), run=130907-130907msec
  WRITE: bw=8026KiB/s (8218kB/s), 8026KiB/s-8026KiB/s (8218kB/s-8218kB/s), io=1026MiB (1076MB), run=130907-130907msec

Disk stats (read/write):
  sda: ios=785092/262566, merge=0/56, ticks=8306368/27570, in_queue=8333947, util=100.00%

Kód: [Vybrat]

fio --randrepeat=1 --ioengine=libaio --direct=1 --gtod_reduce=1 --name=test --filename=random_read_write.fio --bs=4k --iodepth=64 --size=4G --readwrite=randrw --rwmixread=80
test: (g=0): rw=randrw, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=libaio, iodepth=64
fio-3.28
Starting 1 process
test: Laying out IO file (1 file / 4096MiB)
Jobs: 1 (f=1): [m(1)][100.0%][r=23.5MiB/s,w=5804KiB/s][r=6004,w=1451 IOPS][eta 00m:00s]
test: (groupid=0, jobs=1): err= 0: pid=54249: Sun Nov 20 08:12:52 2022
  read: IOPS=6002, BW=23.4MiB/s (24.6MB/s)(3276MiB/139700msec)
   bw (  KiB/s): min=23512, max=28726, per=100.00%, avg=24029.47, stdev=287.11, samples=279
   iops        : min= 5878, max= 7181, avg=6007.37, stdev=71.75, samples=279
  write: IOPS=1503, BW=6014KiB/s (6158kB/s)(820MiB/139700msec); 0 zone resets
   bw (  KiB/s): min= 5320, max= 7696, per=100.00%, avg=6019.15, stdev=272.36, samples=279
   iops        : min= 1330, max= 1924, avg=1504.79, stdev=68.09, samples=279
  cpu          : usr=3.33%, sys=11.72%, ctx=661532, majf=0, minf=7
  IO depths    : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=0.1%, 32=0.1%, >=64=100.0%
     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=838551,210025,0,0 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=23.4MiB/s (24.6MB/s), 23.4MiB/s-23.4MiB/s (24.6MB/s-24.6MB/s), io=3276MiB (3435MB), run=139700-139700msec
  WRITE: bw=6014KiB/s (6158kB/s), 6014KiB/s-6014KiB/s (6158kB/s-6158kB/s), io=820MiB (860MB), run=139700-139700msec

Disk stats (read/write):
  sda: ios=838296/210167, merge=77/109, ticks=8880065/22250, in_queue=8902322, util=100.00%
Pro zajímavost ještě Ubuntu 22.04 (VM na Hyper-V 2022, kde disky jsou v obyčejném Intel RAID1 - 2xS3700)

Kód: [Vybrat]

fio --randrepeat=1 --ioengine=libaio --direct=1 --gtod_reduce=1 --name=test1 --filename=test1 --bs=4k --iodepth=64 --size=4G --readwrite=randrw --rwmixread=75
test1: (g=0): rw=randrw, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=libaio, iodepth=64
fio-3.28
Starting 1 process
test1: Laying out IO file (1 file / 4096MiB)
Jobs: 1 (f=1): [m(1)][100.0%][r=126MiB/s,w=41.6MiB/s][r=32.3k,w=10.7k IOPS][eta 00m:00s]
test1: (groupid=0, jobs=1): err= 0: pid=4139979: Sun Nov 20 07:17:23 2022
  read: IOPS=31.7k, BW=124MiB/s (130MB/s)(3070MiB/24791msec)
   bw (  KiB/s): min=110858, max=144047, per=100.00%, avg=127091.69, stdev=6868.16, samples=49
   iops        : min=27714, max=36011, avg=31772.82, stdev=1717.02, samples=49
  write: IOPS=10.6k, BW=41.4MiB/s (43.4MB/s)(1026MiB/24791msec); 0 zone resets
   bw (  KiB/s): min=37580, max=48662, per=100.00%, avg=42468.63, stdev=2425.38, samples=49
   iops        : min= 9395, max=12165, avg=10617.06, stdev=606.30, samples=49
  cpu          : usr=16.65%, sys=53.03%, ctx=33363, majf=2, minf=6
  IO depths    : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=0.1%, 32=0.1%, >=64=100.0%
     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=785920,262656,0,0 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=124MiB/s (130MB/s), 124MiB/s-124MiB/s (130MB/s-130MB/s), io=3070MiB (3219MB), run=24791-24791msec
  WRITE: bw=41.4MiB/s (43.4MB/s), 41.4MiB/s-41.4MiB/s (43.4MB/s-43.4MB/s), io=1026MiB (1076MB), run=24791-24791msec

Disk stats (read/write):
    dm-0: ios=821137/260866, merge=0/0, ticks=851116/221292, in_queue=1072408, util=100.00%, aggrios=809296/262905, aggrmerge=18687/295, aggrticks=845801/226942, aggrin_queue=1072872, aggrutil=99.87%
  sda: ios=809296/262905, merge=18687/295, ticks=845801/226942, in_queue=1072872, util=99.87%

[SmolPachol]

A tu je ukazkovo vidiet aky podvod contabo to v podstate je a ze nieje hodne aby zato ludia platili ani 1 kč mesacne.

READ: bw=16.3MiB/s (17.1MB/s), 16.3MiB/s-16.3MiB/s (17.1MB/s-17.1MB/s)
WRITE: bw=4171KiB/s (4271kB/s), 4171KiB/s-4171KiB/s (4271kB/s-4271kB/s), io=820MiB (860MB)

16.3MiB/s na read a 4 MiB/s na write je nieco co sa neda akceptovat a akakolvek suma ktoru clovek musi zaplatit je uplne mimo.

Myslim ze jen ne zcela vite co merite. Pro nahodnych 1000 IOPS s velikosti 4k me ta sirka pasma prijde jako ocekavatelna hodnota. Pro jine typy zateze dava disk toho sameho serveru o dva rady vetsi propustnost.

Aj keby som zato dostaval peniaze je to proste nepouzitelne. Akakolvek najlacnejsia SD karta aj USB kluc je omnoho rychlejsie ako tento "virtualny server".

Asi dost zalezi na tom co myslite tim 'rychlejsi'. Neni pravda ze je to nepouzitelne. Realne to pouziva kazdy den rada uzivatelu. Dokazu si predstavit, ze vase aplikace nutne potrebuje o nekolik radu vice IOPS nez kolik vam ji Contabo kdy poskytne. Pak je pro vas ucel nebo aplikaci Contabo urcite nevhodne. Existuje ale mnoho pripadu, pro ktere jsou ta cisla  zcela dostatecna.

Uz cca 12 rokov dozadu som mal SSD v PC. Servre su daleko dolezitejsie a tam nemozu byt pomale disky a nieco co je pomalsie ako SD karta. To je masaker.

Myslim ze je jedno kolik mate SSD, nebo SD v PC. Servery klidne muzou mit disky s mene jak 1000 IOPS. Dokonce i servery vyhrazene jen pro vas.

Dotazujci se ostatne nepta po hostingu s vetsim vykonem, zrejme mu Contabo staci a zrejme i z pohledu propustnosti blokoveho zarizeni. Hleda alternativu v evrope kuli spolehlivosti, respektive zhorsujici se dostupnosti Contabo. Z toho lze usoudit, ze zrejme hleda neco s kratkou sitovou odezvou pro uzivatele v CR.
Moje mereni ani aplikace zvysene vypadky na Contabu v posledni dobe nezaznamenavaji. Presto verim ze je pozoruje. Contabo udrzby vypisuje https://contabo-status.com/ a zrejme mam zase jen stesti a me za posledni rok a pul zadna nezasahla.
A neporadim mu. Bylo by fajn, kdyby nejaky VPS hosting byl super spolehlivy a zaroven taky levny. Ale v case to jde trochu proti sobe.

[Reklama]

[SimisN]

Zdravim,

VPS na Contabo mam cca 2 roky a zatim jsem nemel zadny problem. Sice neprovozuji nic kritickeho, kde bych sledoval dostupnost nepretrzite, ale prvni rok jsem pristupoval nekolikrat denne a nedostupnost jsem nezazil.

Ohledne rychlosti disku

Kód: [Vybrat]


dd if=/dev/zero  of=output.file  bs=1G  count=1 oflag=dsync
1+0 records in
1+0 records out
1073741824 bytes (1.1 GB, 1.0 GiB) copied, 13.9374 s, 77.0 MB/s

dd if=output.file of=/dev/null bs=1M count=1024
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB, 1.0 GiB) copied, 2.81095 s, 382 MB/s


Asi zadny zazrak, ale pro moje ucely staci bohate.