[ikk157]

Greetings everyone!

First off all I would like to apologize for the lack of formatting(e.g. different text sizes, ...) in this review. It’s very hard to do it on mobile and I’m not very familiar on how the text formatting of MyBB works.

So I’ve been using VPS 9 for around a month now, which happens to be the very first VPS that I’ve got to call my own. (Since in the past years, up to the present, I’ve only been setting up/managing the VPSs of my friends and people on the internet who don’t know how to manage their VPSs on their own.) This is a big thing! Why? You may ask... Well, it’s because I’ve never been able to purchase a VPS of my own. That is because I don’t have a credit card, and hence I can’t make online purchases. So the fact that post4vps and virmach generously provided this VPS to me for FREE is something that I still can’t believe to this day! It still feels weird to say “MY VPS”.

First few days with VPS 9

My very first few days weren’t very “pretty” with VPS 9. They involved a bunch of “hiccups” and misunderstandings. The VPS 9 initially came pre-installed with centos. However, I prefer ubuntu. So I requested the OS to be re-installed to ubuntu and things went downhill from there! The VPS stopped working immediately! I requested the OS to be re-installed again. But the outcome was still the same. Eventually due to a massive misunderstanding, the blame was put on me even though I haven’t touched a thing. However, I completely understand how that misunderstanding could’ve taken place, so no one is at fault. Eventually, I managed to clear the misunderstandings and after further testing it was found to be that the ubuntu images virmach had seemed to be broken. So as a last resort, I went with centos which seems to work fine to this day (took a few hours to get used to the minor differences). 


After things were sorted out, I was instantly impressed by how well the VPS performed. It was insanely fast! 

Specs

Now the specs are available on several places on this forum. But since this is a review, it won’t make sense for me not to provide them here as well. So here you go:

Disk space: 100GB
RAM capacity: 8GB
Virtualization technology: KVM
Bandwidth: 4TB
Control Panel: not available 
Connection speed: 1 Gb/s
Provider: Virmach (https://virmach.com)

Benchmarks:

System Info
-----------
Processor       : Intel® Xeon® CPU E5-2670 v2 @ 2.50GHz
CPU Cores       : 2
Frequency       : 2499.998 MHz
Memory          : 7821 MB
Swap            : 7821 MB
Uptime          : 25 days, 17:54,

OS              : CentOS Linux 7 (Core)
Arch            : x86_64 (64 Bit)
Kernel          : 3.10.0-1062.4.1.el7.x86_64
Hostname        : kvm-Post2Host-Seattle


Speedtest
----------

I completely gave up on the network speed test, it was taking forever. (Check the edit at the end of this post)


Disk Speed
----------
I/O (1st run)   : 102 MB/s
I/O (2nd run)   : 331 MB/s
I/O (3rd run)   : 362 MB/s
Average I/O     : 265 MB/s

(Credits to @sohamb03 for the script)

As you can see, apart from the RAM and disk space, the rest of the benchmarks weren’t that impressive. But like, it’s free! So I can’t really complain

I haven’t tried VPS9 back when it was on OVZ. So I can’t make the comparison between that and KVM. However, I don’t like the nature of OVZ, I absolutely HATE how the resources are shared and how your “noisy neighbors” could affect the performance of your VPS, which also resulted into many companies utilizing strict resource policies. However, with KVM, you get your own dedicated resources. Which solves all of the issues that OVZ has.

Pros of VPS 9:

1. Insane amount of RAM
   
2. Insane amount of disk space
   
3. And of course, it’s free!
   

Cons of VPS 9:

1. CPU is not that impressive
   
2. The network speed was slow, hence why the speedtest was taking forever (quite odd considering the 1Gbps shown on the specs).
   
3. No control panel (However, this is something you are made well aware of before applying for the VPS).
   
4. Ubuntu didn’t work as of last month, not sure if they got it fixed now.
   

Would I recommend VPS9?

Well, it really depends on what you’ll use it for. For my use case, it’s the perfect setup! However, people who need more out of the CPU rather than the RAM might not like it as much. Same goes to the ones that need extremely fast network speeds. 

And of course, special thanks to Post4VPS and Virmach for this lovely VPS!

Edit:

I’ve gone with Hidden refuge’s suggestion and used the speedtest cli to run the network speed test. Here’s the outcome:


Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by CenturyLink (Seattle, WA) [13.12 km]: 11.194 ms
Testing download speed................................................................................
Download: 320.31 Mbit/s
Testing upload speed................................................................................................
Upload: 257.47 Mbit/s

I also wanted to go with Geekbench to perform a benchmark. However, Geekbench isn’t free so that’s not an option. So I went on with the second benchmarking program hidden refuge recommended which is Unixbench. Here are the results:



Further edit:

I’ve gone ahead and ran Geekbench according to the suggestion placed by @fChk.

You can find the results here: https://browser.geekbench.com/v5/cpu/2077291

[deanhills]

Excellent review @ikk157!  Enjoyed reading it.  You have a gift of the gab as they say.  A way with words.

Am very happy to hear that the VPS performed OK from the moment CentOS was loaded.  And apologies for the misunderstanding before.

If I may ask - did you disable Selinux before you started to use CentOS 7, or did everything work straight off the bat?  In other words did you need to make any modifications before you were able to use CentOS successfully?

Your VPS 9 (Seattle) was one of the first two VPSs that the technicians at Virmach upgraded from OpenVZ 6 to KVM, hence why I am asking the question, because looks as though they must have done a real good job with the first 2 post4vps VPS 9 upgrades - Seattle and Chicago, whereas in later upgrades of the other 6 VPSs, they started to skip steps to speed the process. Mine was the very last VPS 9 that got upgraded, and I've been having serious issues with it.

I'm particularly thinking of what @fChk had commented in another thread that if the sponsor had properly put the upgraded VPSs together the issue with Selinux should not have occurred.  It is a sign that possibly the reconfiguration of the later upgrades didn't go as detailed and with the same care as the first two VPSs - Chicago and Seattle had been.

In the meanwhile I'm happy VPS 9 Seattle is performing well in spite of not having been able to work with Ubuntu on it.

[ikk157]

Excellent review @ikk157!  Enjoyed reading it.  You have a gift of the gab as they say.  A way with words.

Am very happy to hear that the VPS performed OK from the moment CentOS was loaded.  And apologies for the misunderstanding before.

If I may ask - did you disable Selinux before you started to use CentOS 7, or did everything work straight off the bat?  In other words did you need to make any modifications before you were able to use CentOS successfully?

Your VPS 9 (Seattle) was one of the first two VPSs that the technicians at Virmach upgraded from OpenVZ 6 to KVM, hence why I am asking the question, because looks as though they must have done a real good job with the first 2 post4vps VPS 9 upgrades - Seattle and Chicago, whereas in later upgrades of the other 6 VPSs, they started to skip steps to speed the process.  Mine was the very last VPS 9 that got upgraded, and I've been having serious issues with it.

I'm particularly thinking of what @fChk had commented in another thread that if the sponsor had properly put the upgraded VPSs together the issue with Selinux should not have occurred.  It is a sign that possibly the reconfiguration of the later upgrades didn't go as detailed and with the same care as the first two VPSs - Chicago and Seattle had been.

In the meanwhile I'm happy VPS 9 Seattle is performing well in spite of not having been able to work with Ubuntu on it.

Thank you for the kind words! I highly appreciate that you find my review good.

As for your question:

Nope! I didn’t have to do anything at all! Centos worked perfectly “out of the box”. I’ve seen the complaints regarding the other VPSs experiencing issues after the KVM upgrade. The one with the decreased storage specially caught my attention. It’s quite odd in my opinion that Virmach is doing the upgrades manually. They should’ve automated the whole process to speed things up and get more consistent results. Correct me if I’m wrong.

[Mashiro]

Great review so far. What a bummer to hear that the network speed was so low. Maybe an alternative way to take a few speed tests for you is to use the Speedtest.net CLI client and perform tests from various different servers (locations)? I'm curious to know why the speed is so bad.

Official CLI app: https://www.speedtest.net/apps/cli
Speedtest-CLI by Sivel: https://github.com/sivel/speedtest-cli


The CPU used is clearly not made for highest peformance per core but rather for multicore applications or hosting of many VMs that don't require high clocks. Generally though the E5s are great CPUs. Many cores and threads plus usually they are dual socket systems meaning dual amount of cores and threads. If a higher performance is necessary one is better off with a Intel Xeon E3 though due to its much higher clock speeds but you pay with reduced amount of cores, threads and less overall RAM support.

To actually show off how well/bad the CPU is a few benchmarks are necessary. For such things there is software like Geekbench or Byte Unixbench. This is something that could be added to actually give some value to the review and the statement about the CPU.

[ikk157]

Great review so far. What a bummer to hear that the network speed was so low. Maybe an alternative way to take a few speed tests for you is to use the Speedtest.net CLI client and perform tests from various different servers (locations)? I'm curious to know why the speed is so bad.

Official CLI app: https://www.speedtest.net/apps/cli
Speedtest-CLI by Sivel: https://github.com/sivel/speedtest-cli


The CPU used is clearly not made for highest peformance per core but rather for multicore applications or hosting of many VMs that don't require high clocks. Generally though the E5s are great CPUs. Many cores and threads plus usually they are dual socket systems meaning dual amount of cores and threads. If a higher performance is necessary one is better off with a Intel Xeon E3 though due to its much higher clock speeds but you pay with reduced amount of cores, threads and less overall RAM support.

To actually show off how well/bad the CPU is a few benchmarks are necessary. For such things there is software like Geekbench or Byte Unixbench. This is something that could be added to actually give some value to the review and the statement about the CPU.

Thank you for the great feedback!

I have gone ahead and added in the speedtest-cli and unixbench results (check the edit). I couldn’t use Geekbench unfortunately since it’s a paid program. So I went with unixbench.

---

I just noticed that the image showing the unixbench results isn’t loading up within the post at all. It’s only showing the image tag and the url of the image. Not sure how to fix this. Would someone please help me with that?

[fChk]

I was planning to do a meta-review of Virmach VPS-9 offerings -especially CPU wise- (a review of reviews), thus when I was going through the available threads on the subject, it appeared that @ikk157 was the only one who did run a Unixbench test :-)

Not only that, he did it in double-test mode!.. Kudos! and +1Rep for your courage :-)

Did you know that you most probably broke Virmarsh's ACCEPTABLE USAGE POLICY while doing it ? Both the high CPU and high load clauses:

High CPU: Customer’s Service cannot burst to 95-100% usage for more than five (5) minutes and cannot average higher than 50% usage within any two (2) hour period. Packages advertised to include dedicated CPU, Services with the high CPU option, and any customized Service plans that include high CPU option may burst to 100% at all times.

High Load: Customer’s Service cannot have a 15-minute load average higher than the number of full logical cores assigned and cannot have a 1-day load average higher than 70% of the number of full logical cores assigned.

I knew that for a fact because I was planning to run both the simple and double tests my self on the Phoenix-based VPS, but ended up doing the simple test only as it was the only one who can finish within Virmarch's rules, ie a load of 6.18 (1') 3.65 (5') 1.88 (15') and a CPU usage of 56% for most of the test duration and ending at 95.83% .

For the meta-review thing I did cancel it as there is not enough data in the published reviews for KVM-VPS-9 while others are still missing!

I also wanted to go with Geekbench to perform a benchmark. However, Geekbench isn’t free so that’s not an option. So I went on with ....

You can still use it in its trial mode-like everybody I would guess and it's more than enough. By the way, it doesn't abuse the CPU the way UnixBench does.

I haven’t tried VPS9 back when it was on OVZ. So I can’t make the comparison between that and KVM. However, I don’t like the nature of OVZ, I absolutely HATE how the resources are shared and how your “noisy neighbors” could affect the performance of your VPS, which also resulted into many companies utilizing strict resource policies. However, with KVM, you get your own dedicated resources. Which solves all of the issues that OVZ has.

The “noisy neighbours” effect can still be felt within KVM's VMs; it shows up CPU-wise in the steal time statistic when querying your CPU usage.

When the steal time gets above 10%, your vCPU's performance drop as it's struggling to get its fair share of the real CPU from the hypervisor.

Container's VM (like OVZ) vCPU performance suffers most from the IOWait time statistic which can jump to levels above 30% making everything crawl to a halt.

It's all relative, and the main culprit in both cases is usually the overselling of servers' resources.

[youssefbasha]

Awesome review!
About the speedtest i think that you can open a topic for it but for my side i see it acceptable.

[ikk157]

I was planning to do a meta-review of Virmach VPS-9 offerings -especially CPU wise- (a review of reviews), thus when I was going through the available threads on the subject, it appeared that @ikk157 was the only one who did run a Unixbench test :-)

Not only that, he did it in double-test mode!.. Kudos! and +1Rep for your courage :-)

Did you know that you most probably broke Virmarsh's ACCEPTABLE USAGE POLICY while doing it ? Both the high CPU and high load clauses:


I knew that for a fact because I was planning to run both the simple and double tests my self on the Phoenix-based VPS, but ended up doing the simple test only as it was the only one who can finish within Virmarch's rules, ie a load of 6.18 (1') 3.65 (5') 1.88 (15') and a CPU usage of 56% for most of the test duration and ending at 95.83% .

For the meta-review thing I did cancel it as there is not enough data in the published reviews for KVM-VPS-9 while others are still missing!

You can still use it in its trial mode-like everybody I would guess and it's more than enough. By the way, it doesn't abuse the CPU the way UnixBench does.



The “noisy neighbours” effect can still be felt within KVM's VMs; it shows up CPU-wise in the steal time statistic when querying your CPU usage.

When the steal time gets above 10%, your vCPU's performance drop as it's struggling to get its fair share of the real CPU from the hypervisor.

Container's VM (like OVZ) vCPU performance suffers most from the IOWait time statistic which can jump to levels above 30% making everything crawl to a halt.

It's all relative, and the main culprit in both cases is usually the overselling of servers' resources.

VirMach’s High CPU policy was back when the VPSs were on OVZ6. Since with OVZ, if you burst into a significantly high CPU usage you will affect your “neighbors”.

With KVM, on the other hand, it doesn’t really affect your neighbors much at all (sure, just like you’ve described, there is still a minor effect).

But again, the benchmark only ran for an extremely short amount of time so things should be fine. 

Would love to have a clarification regarding high CPU usage from an admin now that VPS 9 is a KVM.

As for Unixbench, thank you very much for your great comment! I wanted to avoid Geekbench altogether as the trial mode is not as in depth... Unixbench should do the trick.

---

Awesome review!
About the speedtest i think that you can open a topic for it but for my side i see it acceptable.

I don’t think the admins can really do anything regarding the low network speeds (for a matter of fact, I noticed that they’ve actually worsened now).

I’ve already bothered the admins enough and this just isn’t something I’d want to have them spend time on investigating... not to mention that all of this is free so we really don’t have the right to complain!

---

Update regarding the network speed on my VPS 9:

When running speedtest-cli, I get the following results:

Download: 873.11 Mbit/s
Upload: 1051.91 Mbit/s

As you can see, the network speeds here are excellent.

However, when actually downloading something (let’s say for example when running apt upgrade), it seems to struggle!

Even connecting to it through ssh takes longer than any of the other VPSs I’ve used.

I’m starting to suspect that the SSD is the bottleneck here, not the network speed. I’m not very sure though.

[fChk]

VirMach’s High CPU policy was back when the VPSs were on OVZ6. Since with OVZ, if you burst into a significantly high CPU usage you will affect your “neighbors”.

With KVM, on the other hand, it doesn’t really affect your neighbors much at all (sure, just like you’ve described, there is still a minor effect).

No!.. It's not a minor effect!.. CPU contention between VMs is as real as it is between Containers, as I've tried to explain previously.

But again, the benchmark only ran for an extremely short amount of time so things should be fine.

That's not what your posted image says!.. ie 28m 17sec.

Unixbench is a system benchmark (not CPU only) and it generally take 20+ minutes, well above the 5min limitation.

Given that I've never heard of Virmach before joining this forum, I'm not quite sure of their reputation and how well they enforce their AUP, if at all. But I would rather stay on the safe side of the bargain.

As for Unixbench, thank you very much for your great comment! I wanted to avoid Geekbench altogether as the trial mode is not as in depth... Unixbench should do the trick.

The point is that Geekbench in trial mode was USED in other reviews, hence its scores should be another point of comparisons between VPS's CPUs. That's all.

[ikk157]

No!.. It's not a minor effect!.. CPU contention between VMs is as real as it is between Containers, as I've tried to explain previously.

That's not what your posted image says!.. ie 28m 17sec.

Unixbench is a system benchmark (not CPU only) and it generally take 20+ minutes, well above the 5min limitation.

Given that I've never heard of Virmach before joining this forum, I'm not quite sure of their reputation and how well they enforce their AUP, if at all. But I would rather stay on the safe side of the bargain.


The point is that Geekbench in trial mode was USED in other reviews, hence its scores should be another point of comparisons between VPS's CPUs. That's all.

I stand corrected... never knew it affected the CPU that much even in KVM. My entire idea of KVM was that your resources are yours and hence you using 100% of them doesn’t make a difference to anyone else.

I just realized how long the benchmark actually ran for... woah never noticed it actually took that long... thanks for the correction.

I’ll go ahead and run the trial mode of Geekbench and add it to my review in a few hours. 

Thank you for the clarification and quality feedback on the review!

---

Update: I’ve gone ahead and ran Geekbench then added it to the review.
[url=https://browser.geekbench.com/v5/cpu/2077291][/url]