The "4GB for $7" match: DotVPS vs INIZ (OpenVZ.io)

[Tracid]

Scenario

I wanted to rent a Kimsufi to play around with my dusty sysadmin skills. In this case, it always comes in handy to have the possibility to quickly reinstall the OS or try another distribution. Since it's just a sandbox, I don't need scaling in the cloud. I live in Europe, so let's have it in a local datacenter, in favor of lower latency and faster speeds. What about a VPS? Sure, because I am not storing any sensitive data in case of a jailbreak.

Offers

DotVPS: London, 4GB RAM, 4 vCore, 100GB disk space, 1500GB bandwidth, $25 discounted to, $7 recurring

INIZ (OpenVZ.io): Amsterdam, 4GB RAM, 3 vCore, 75GB disk space, 1000GB bandwidth, $9.36 discounted 25% off to $7.49, recurring

Winner: Draw! INIZ offers less disk space but has unmetered incoming bandwidth in change (DotVPS measures both outgoing and incoming). See the tests below on comparison regarding the vCores.

Ordering & control panel

Winner: INIZ! Both companies use the same SolusVM platform, so they aren't really any functional differences. But there's a minor difference: INIZ activates your service right after ordering, before payment has been received, while DotVPS activates it in less than 5 minutes after payment. Also, DotVPS doesn't allow you to set the reverse DNS from the control panel, but that is resolved fast thru a support ticket.

Support

INIZ: is represented by Patrick @ vpsBoard

DotVPS: is represented by Jack @ vpsBoard

Issue resolving

Both of the providers were missing Ubuntu 13.10 64 bit from their OS template list, so I requested them to add it. The tickets were submitted as lowest priority, on sunday evening (UTC).

DotVPS: request fulfilled in 6 hours, free of charge.

INIZ: request fulfilled in 27 hours, free of charge. As a bonus, they also added the 32 bit version, and apologized for the delay.

Winner: DotVPS! The delay of INIZ could be explained by timezone difference & weekend, but DotVPS did great even on Sunday.

First round winner: Draw! The following post will cover the tests.

 

[Tracid]

Based on wlanboy's review tests:

CPU

INIZ: (3x)

processor : 0
vendor_id : GenuineIntel
cpu family : 6
model : 45
model name : Intel(R) Xeon(R) CPU E5-2650 0 @ 2.00GHz
stepping : 7
cpu MHz : 1995.361
cache size : 20480 KB
physical id : 0
siblings : 16
core id : 0
cpu cores : 8
apicid : 0
initial apicid : 0
fpu : yes
fpu_exception : yes
cpuid level : 13
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good xtopology nonstop_tsc aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes xsave avx lahf_lm ida arat epb xsaveopt pln pts dts tpr_shadow vnmi flexpriority ept vpid
bogomips : 3990.72
clflush size : 64
cache_alignment : 64
address sizes : 46 bits physical, 48 bits virtual
power management:
DotVPS: (4x)

processor	: 0
vendor_id	: GenuineIntel
cpu family	: 6
model		: 44
model name	: Intel(R) Xeon(R) CPU           X5650  @ 2.67GHz
stepping	: 2
cpu MHz		: 2667.012
cache size	: 12288 KB
physical id	: 0
siblings	: 12
core id		: 0
cpu cores	: 6
apicid		: 0
initial apicid	: 0
fpu		: yes
fpu_exception	: yes
cpuid level	: 11
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good xtopology nonstop_tsc aperfmperf pni dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm pcid dca sse4_1 sse4_2 popcnt lahf_lm ida arat epb dts tpr_shadow vnmi flexpriority ept vpid
bogomips	: 5334.02
clflush size	: 64
cache_alignment	: 64
address sizes	: 40 bits physical, 48 bits virtual
power management:

Winner: Draw! INIZ offers 3 vCores of a newer, more performant* CPU, compared to the 4 vCores offered by DotVPS. (* according to CPU benchmark list)

Disk performance

DotVPS:

dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync && rm -rf test
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 2.69423 s, 399 MB/s
INIZ:

dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync && rm -rf test
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 2.1292 s, 504 MB/s

Winner: INIZ!

Network:

INIZ:

wget cachefly.cachefly.net/100mb.test -O /dev/null
--2014-03-04 18:35:41-- http://cachefly.cachefly.net/100mb.test
Resolving cachefly.cachefly.net (cachefly.cachefly.net)... 205.234.175.175
Connecting to cachefly.cachefly.net (cachefly.cachefly.net)|205.234.175.175|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 104857600 (100M) [application/octet-stream]
Saving to: '/dev/null'

100%[====================================================================================================================================================================================================================================>] 104,857,600 63.5MB/s in 1.6s

2014-03-04 18:35:42 (63.5 MB/s) - '/dev/null' saved [104857600/104857600]
DotVPS:

wget cachefly.cachefly.net/100mb.test -O /dev/null
--2014-03-04 18:35:44--  http://cachefly.cachefly.net/100mb.test
Resolving cachefly.cachefly.net (cachefly.cachefly.net)... 205.234.175.175
Connecting to cachefly.cachefly.net (cachefly.cachefly.net)|205.234.175.175|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 104857600 (100M) [application/octet-stream]
Saving to: '/dev/null'

100%[====================================================================================================================================================================================================================================>] 104,857,600  101MB/s   in 1.0s   

2014-03-04 18:35:46 (101 MB/s) - '/dev/null' saved [104857600/104857600]

Winner: DotVPS!

Overall winner: Draw! You need to make a choice depending on your needs.

 

[manacit]

To be fair, one wget to cachefly doesn't represent a fraction of the actual network, and the E5 is a significantly newer processor than the X series Xeon that DotVPS has. 

 

[drmike]

Meh, two more Mc-OpenVZ offers...   With oversell opportunity about to be realized...  Have to maximize profits...

Disk speed looks good on both.  Slews of drives and hardware RAID does that.

I'd like to see some extended use numbers and CPU performance.

I'm not a mathematician, but then again neither are those exploiting OpenVZ   $7 for 4GB of RAM...   Even if oh, a 384GB server = 96 1-1 RAM to sold RAM plans....  

96 x 3 ratio for oversell = 288 

288 x 7 = $2016

Looks like good coin but a 384GB RAM server isn't cheap.   If using less RAM.. well numbers drop to... ahhh

192GB / 4GB = 48 acccounts

48 x 3 ratio for oversell = 144

144 x $7 = $1008....

So....  144 - 288 accounts... $1-2k income pre all costs...   

Guess I could price these servers with 12 drives and come close to figuring out the riddle.

Clearly DotVPS model on the older hardware is superior business wise where able to buy gear and own. 

I remain frightened by the many workloads being packed into either server... With newer gear, encouraging folks to get more wild in loading as notably the CPUs are more capable of abuse.

 

[Tracid]

@manacit: both disk & network tests were ran 10 times and I've selected the most average result.

 

[Tracid]

DotVPS update:

DotVPS have upgraded their servers and are now using Dual E5-2630(v2) CPU's.

They've been experiencing some problems caused by faulty hardware, that's the reason why users were migrated to an older node (like the one I've ran the tests on) to prevent service interruption.

This puts DotVPS on step ahead INIZ.

But, as an extra bonus for being temporarily migrated to an "older" server, users got 1 month of free credit.

A huge +1 for client appreciation!

 

[Tracid]

DotVPS update:

We are experiencing DDoS attacks on all of our IP subnets in the UK location. They are all nullrouted by the datacenter for now. There is no ETA to get everything back online for now.

...since over 3 days already.

I guess it's the datacenter to blame, and not the VPS provider.

 

[serverian]

DotVPS update:


...since over 3 days already.


I guess it's the datacenter to blame, and not the VPS provider.

Please open a ticket to get updates.

 

[Tracid]

Update:

I am still very happy with both providers.

@Patrick of INIZ is providing great support.

DotVPS is now undergoing a transition, it's becoming Virtora. Even though @Jack left, Oktay a.k.a. @serverian still provides great support.

To be honest, I don't need two of these services, but I can't decide which one to give up. If things will still go so smooth, I'll keep both for the upcoming years.

 

[Tracid]

Update:

Virtora is still offering the same great service as DotVPS did.

INIZ has upgraded the hardware to E5-2650v2 CPUs with a really short downtime.

 

[willie]

I don't see the point of these ridiculous oversold plans.  You can't use them like a dedi.  There's nothing like kicking off an 8-hour multicore 100% cpu task at bedtime, checking the results the next morning and seeing that they look good but misformatted, and instead of spending 15 minutes massaging the output into the right shape by hand, you spend 2 seconds fixing the script and then launch the whole 8 hour cpu-intensive task again before you leave for work.  You'll get thrown off of any budget VPS host for doing that, but with a dedi it's no sweat at all.   Or try running a low traffic Redis database on that 4gb VPS.  Not much CPU needed, but there will be a big speed hit if the data gets swapped out of real ram, which is sure to happen with the VPS.  If you really need a 4GB plan and if a kimsufi dedi fills your requirements, get the dedi.  Main thing missing from them are range of locations, and RAID storage.