HostLease Reviewing my VPS 16 by Hostlease

Hello Post4VPS Community & Guests

During the last giveaway of 2019, in December, I applied for the VPS 16 and won the VPS during the giveaway process. In about three to four days a whole month of ownership and usage will be reached for me with the VPS 16. Unfortunately in about exactly the same amount of time my vacation will end and after that I won't have much time to do a lot on it due to my real life job. So I decided to write a "after a month of usage" review to share my experience with the VPS so far before I have no time or totally forget about it.

I should note that this is my very first Post4VPS VPS since I joined back in Q3 of 2015. I have not owned a VPS here before and neither did I review any VPS or other services here before. This also will be the first review for the VPS 16 as it is actually a brand new VPS that was in the giveaway process for the first time so far (CMIIW).

Let's get started.

VPS Specifications

First I would like to list the VPS specifications here again so you don't have to look it up somewhere or have to guess it. It will be easier to have all information on a single page when reading a review to quickly look things up if necessary. I'll also confirm the specifications by reading the information directly from the VPS using a bash benchmarking script.

Specifications
CPU: Intel Xeon E5-2620 v2 @ 2.10 GHz
CPU Cores: 4
RAM: 12 GB
Disk Space: 200 GB (SSD)
Bandwidth: 5 TB / month
Connection: 1 Gbit/s
IP Adress(es): 1x IPv4 Address
Location: Lelystad (close to Amsterdam) in the Netherlands
Virtualization: KVM (Hardware virtualization)
Control Panel: None provided (support via the support forums)

I can confirm this VPS specifications via my bench.sh 2.0 benchmarking script. The information is available below.

System Info

Code:
Processor       : Intel(R) Xeon(R) CPU E5-2620 v2 @ 2.10GHz

CPU Cores       : 4 @ 2095 MHz

Memory          : 12010 MB

Swap            : 0 MB

Uptime          : 25 days, 20:45,

OS              : Debian GNU/Linux 10 (buster)

Arch            : x86_64 (64 Bit)

Kernel          : 4.19.0-6-amd64

Hostname        : <hostname redacted>

As you can see the specifications check out with the information provided by Hostlease on the VPS plans page. Disk space also checks out (see information listed below).

Disk Space

Code:
Filesystem      Size  Used Avail Use% Mounted on

udev            5.9G     0  5.9G   0% /dev

tmpfs           1.2G   16M  1.2G   2% /run

/dev/vda1       197G  2.2G  195G   2% /

tmpfs           5.9G     0  5.9G   0% /dev/shm

tmpfs           5.0M     0  5.0M   0% /run/lock

tmpfs           5.9G     0  5.9G   0% /sys/fs/cgroup

tmpfs           1.2G     0  1.2G   0% /run/user/0

VPS Usage

Before getting into checking and benchmarking the performance of the VPS I would like to mention how and for what I will be using the VPS. Due to the high specifications of the VPS the giveaway process is not so simple as you might guess. It is necessary to list usage and reason why you should get the VPS instead of everyone else. Basically not only the your score matters but also "how you sell yourself" while applying to win the VPS.

I'm using Debian "Buster" 10 as my OS on the VPS and the main usage of the VPS is a Docker environment to containerize all other applications I want to host such as:

Private NextCloud instance for myself and a few friends (Docker).
Web server for all of my domains and a few websites I would like to run (Docker).
Development, testing and improvement of new and existing scripts.
Self-hosted pastebin service with CLI upload support like termbin (Docker).
Using the VPS resources to write tutorials for this forum and help users with issues or requests.
Maybe a private game server for different games to play on with friends if I find time and people to play with (Docker).
A private VPN server for myself as my NordVPN subscription will run out in four months and I won't be extending it (Docker).
Additionally I would most likely use left over resources to testing new stuff.

Below are a few resources and guides that I've already written and tested on my VPS 16 as a part of my usage plans:

Easy way to upgrade Debian 9 to Debian 10 as Hostlease only had Debian 9 and I wanted to use Debian 10.
Disable Swap and Regain Storage Space on Linux because I don't need Swap with 12 GB RAM and would rather have more usable disk space.
Installing Docker & Portainer on Debian 10 in order to get started with Docker (this series will be continued once I actually find time and also expand my Docker knowledge).

As you can see I didn't manage to do that much so far. Finding free time is simply the main issue. I worked through half of December. And then I had vacation which was mixed with a lot of holidays due to Christmas and New Year. I have the base setup already with Debian 10 and Docker + Portainer installed. I also actually already run a Nginx Docker instance with automated LetsEncrypt DNS wildcard renewal through the CloudFlare API (a guide for this will follow). Additionally I used the VPS already to test my bash VPS benchmarking script and develop new features in the dev branch.

That's it about my usage plans for the VPS and how I will continue with the VPS in the future.

VPS Performance Benchmarks

Now I would like to review the actual performance of the VPS and provide data that will backup my claims and statements. I'll start with the CPU performance, followed by the disk (SSD) performance and finally the network speed.

CPU Performance

The CPU (Intel Xeon E5-2620 v2) used in the physical server is a bit dated for todays standards as it was released in Q3 of 2013. That's around 7 years ago by now. So you cannot expect top notch and up to mark performance. So far I can tell that I couldn't feel any big slowdown when doing my normal tasks. The actual really long taking and slow task was one of the benchmarks. However the benchmark represents the age and performance loss VS newer CPUs well enough to justify the long time it needed to run. Despite having 4 cores you have to always remember that you're not alone on the KVM hypervisor and that there are always other customers also making use of their resources. So in my opinion the performance is fair enough for the age of the CPU and so far I had no issues doing what I needed to do.

Data to show the CPU performance is supplied by a Geekbench 5 and a Byte Unixbench benchmark available below.

Byte Unixbench (Single Core): 414.5 points

Code:
Benchmark Run: Wed Jan 01 2020 11:32:21 - 12:00:27

4 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       18998776.7 lps   (10.0 s, 7 samples)

Double-Precision Whetstone                     3154.6 MWIPS (8.9 s, 7 samples)

Execl Throughput                               1503.7 lps   (30.0 s, 2 samples)

File Copy 1024 bufsize 2000 maxblocks        211466.9 KBps  (30.0 s, 2 samples)

File Copy 256 bufsize 500 maxblocks           66314.3 KBps  (30.0 s, 2 samples)

File Copy 4096 bufsize 8000 maxblocks        624771.7 KBps  (30.0 s, 2 samples)

Pipe Throughput                              472633.0 lps   (10.0 s, 7 samples)

Pipe-based Context Switching                  14045.9 lps   (10.0 s, 7 samples)

Process Creation                               2015.3 lps   (30.0 s, 2 samples)

Shell Scripts (1 concurrent)                   2633.6 lpm   (60.0 s, 2 samples)

Shell Scripts (8 concurrent)                    675.7 lpm   (60.1 s, 2 samples)

System Call Overhead                         343866.1 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX

Dhrystone 2 using register variables         116700.0   18998776.7   1628.0

Double-Precision Whetstone                       55.0       3154.6    573.6

Execl Throughput                                 43.0       1503.7    349.7

File Copy 1024 bufsize 2000 maxblocks          3960.0     211466.9    534.0

File Copy 256 bufsize 500 maxblocks            1655.0      66314.3    400.7

File Copy 4096 bufsize 8000 maxblocks          5800.0     624771.7   1077.2

Pipe Throughput                               12440.0     472633.0    379.9

Pipe-based Context Switching                   4000.0      14045.9     35.1

Process Creation                                126.0       2015.3    159.9

Shell Scripts (1 concurrent)                     42.4       2633.6    621.1

Shell Scripts (8 concurrent)                      6.0        675.7   1126.2

System Call Overhead                          15000.0     343866.1    229.2

                                                                   ========

System Benchmarks Index Score                                         414.5

Byte Unixbench (Multi Core): 1058.8 points

Code:
Benchmark Run: Wed Jan 01 2020 12:00:27 - 12:28:54

4 CPUs in system; running 4 parallel copies of tests

Dhrystone 2 using register variables       63036166.1 lps   (10.0 s, 7 samples)

Double-Precision Whetstone                    11806.1 MWIPS (9.0 s, 7 samples)

Execl Throughput                               3846.6 lps   (29.9 s, 2 samples)

File Copy 1024 bufsize 2000 maxblocks        258705.1 KBps  (30.0 s, 2 samples)

File Copy 256 bufsize 500 maxblocks           72840.8 KBps  (30.0 s, 2 samples)

File Copy 4096 bufsize 8000 maxblocks        730975.8 KBps  (30.0 s, 2 samples)

Pipe Throughput                             1709591.5 lps   (10.0 s, 7 samples)

Pipe-based Context Switching                 200764.4 lps   (10.0 s, 7 samples)

Process Creation                               4778.4 lps   (30.0 s, 2 samples)

Shell Scripts (1 concurrent)                   5771.5 lpm   (60.0 s, 2 samples)

Shell Scripts (8 concurrent)                   1109.2 lpm   (60.1 s, 2 samples)

System Call Overhead                        1203340.6 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX

Dhrystone 2 using register variables         116700.0   63036166.1   5401.6

Double-Precision Whetstone                       55.0      11806.1   2146.6

Execl Throughput                                 43.0       3846.6    894.5

File Copy 1024 bufsize 2000 maxblocks          3960.0     258705.1    653.3

File Copy 256 bufsize 500 maxblocks            1655.0      72840.8    440.1

File Copy 4096 bufsize 8000 maxblocks          5800.0     730975.8   1260.3

Pipe Throughput                               12440.0    1709591.5   1374.3

Pipe-based Context Switching                   4000.0     200764.4    501.9

Process Creation                                126.0       4778.4    379.2

Shell Scripts (1 concurrent)                     42.4       5771.5   1361.2

Shell Scripts (8 concurrent)                      6.0       1109.2   1848.7

System Call Overhead                          15000.0    1203340.6    802.2

                                                                   ========

System Benchmarks Index Score                                        1058.8

Geekbench
Single Core: 292 points
Multi Core: 742 points
Results: https://browser.geekbench.com/v5/cpu/921306

A little side node: I do experience steal of up to 2% per core when checking the core load with advanced details using htop. The CPU steal is probably caused by noise of the other VPS containers. However 2% steal are absolutely ok and not a problem so far.

The benchmarks do confirm my experience so far. The results for single core applications are low but fair. The multi core experience is much better and I can feel it when using applications that actually support multiple cores and use them. That might contribute lot to why I haven't had many bad or really slow experiences so far. Overall the E5 series of Xeon CPUs is exactly meant for this purpose: multi core tasks that don't require too much performance per core. I can however imagine that the CPU will most likely make the game server experience rather difficult as most game servers such as Minecraft or CSGO are performing a lot better and high clocking single cores and don't properly or fully utilize multiple cores. Time will tell.

SSD Performance

Unfortunately I have no clue which SSDs, how many SSDs and in which RAID setup the provider Hostlease is using in that server. I cannot really even make any assumptions or similar because we have zero information on that. So instead I will report about my experience and provide several different benchmark results as crunching data.

dd Performance (three runs)

Code:
I/O (1st run)   : 80.7 MB/s

I/O (2nd run)   : 170 MB/s

I/O (3rd run)   : 159 MB/s

Average I/O     : 136 MB/s or .13 GB/s

(Data supplied by bench.sh 2.0 benchmark script.)

hdparm Speed Test (Cached)

Code:
Timing cached reads:   7358 MB in  1.99 seconds = 3688.36 MB/sec

Timing buffered disk reads: 1016 MB in  3.01 seconds = 337.82 MB/sec

(Data supplied by hdparm tool.)

hdparm Speed Test (Direct)

Code:
Timing O_DIRECT cached reads:   1198 MB in  2.01 seconds = 596.74 MB/sec

Timing O_DIRECT disk reads: 974 MB in  3.00 seconds = 324.30 MB/sec

(Data supplied by hdparm tool.)

Disk Latency (ioping)

Code:
KiB <<< . (ext4 /dev/vda1): request=1 time=435.9 us (warmup)

KiB <<< . (ext4 /dev/vda1): request=2 time=1.07 ms

KiB <<< . (ext4 /dev/vda1): request=3 time=553.6 us

KiB <<< . (ext4 /dev/vda1): request=4 time=2.31 ms

KiB <<< . (ext4 /dev/vda1): request=5 time=3.56 ms

KiB <<< . (ext4 /dev/vda1): request=6 time=5.10 ms

KiB <<< . (ext4 /dev/vda1): request=7 time=17.5 ms (slow)

KiB <<< . (ext4 /dev/vda1): request=8 time=795.7 us (fast)

KiB <<< . (ext4 /dev/vda1): request=9 time=3.31 ms

KiB <<< . (ext4 /dev/vda1): request=10 time=929.9 us (fast)

KiB <<< . (ext4 /dev/vda1): request=11 time=926.6 us (fast)

KiB <<< . (ext4 /dev/vda1): request=12 time=797.5 us (fast)

KiB <<< . (ext4 /dev/vda1): request=13 time=7.90 ms

KiB <<< . (ext4 /dev/vda1): request=14 time=7.98 ms

KiB <<< . (ext4 /dev/vda1): request=15 time=1.49 ms

KiB <<< . (ext4 /dev/vda1): request=16 time=681.0 us (fast)

KiB <<< . (ext4 /dev/vda1): request=17 time=674.3 us (fast)

KiB <<< . (ext4 /dev/vda1): request=18 time=4.27 ms

KiB <<< . (ext4 /dev/vda1): request=19 time=779.5 us (fast)

KiB <<< . (ext4 /dev/vda1): request=20 time=753.2 us (fast)

--- . (ext4 /dev/vda1) ioping statistics ---

requests completed in 61.4 ms, 76 KiB read, 309 iops, 1.21 MiB/s

generated 20 requests in 19.0 s, 80 KiB, 1 iops, 4.21 KiB/s

min/avg/max/mdev = 553.6 us / 3.23 ms / 17.5 ms / 4.07 ms

(Data supplied by the ioping tool.)

Disk Seek Rate (ioping)

Code:
--- /dev/vda (block device 200 GiB) ioping statistics ---

3.02 k requests completed in 2.95 s, 11.8 MiB read, 1.02 k iops, 4.00 MiB/s

generated 3.02 k requests in 3.00 s, 11.8 MiB, 1.01 k iops, 3.93 MiB/s

min/avg/max/mdev = 95.1 us / 975.4 us / 61.2 ms / 3.24 ms

(Data supplied by the ioping tool.)

Disk Sequential Speed (ioping)

Code:
--- /dev/vda (block device 200 GiB) ioping statistics ---

1.65 k requests completed in 2.95 s, 412.8 MiB read, 559 iops, 139.9 MiB/s

generated 1.65 k requests in 3.01 s, 413 MiB, 549 iops, 137.3 MiB/s

min/avg/max/mdev = 361.5 us / 1.79 ms / 62.8 ms / 3.72 ms

(Data supplied by the ioping tool.)

As said before I don't know anything about the SSD setup on the physical server. The disk speed here seems to be fair and enough to work with. The disk latency is also ok although there seem to be little spikes in usage causing slowdown. Overall I didn't feel or see any issues, yet. So all in all I would say it is absolutely ok at the moment although you can generally expect more of SSDs (but again the setup is unknown and we don't even know how many other containers there are and what they're doing). Cannot really complain at that point.

Network Performance

I tested the network performance using wget to download big files from several different servers around the world. I tested IPv4 and IPv6. Yes, despite the VPS plan only saying that the VPS has one IPv4 address I actually noticed that mine also had a public IPv6 address already assigned. In additional I also performed a speedtest.net CLI speed test and did some iperf3 speed tests.

wget Downstream (IPv4)

Code:
Location                Provider        Speed           Latency

Global

CDN                     Cachefly        25.6MB/s        2.904 ms

United States

Atlanta, GA, US         Coloat          5.45MB/s        92.722 ms

Dallas, TX, US          Softlayer       4.50MB/s        116.795 ms

Seattle, WA, US         Softlayer       4.14MB/s        139.033 ms

San Jose, CA, US        Softlayer       4.96MB/s        144.838 ms

Washington, DC, US      Leaseweb        8.24MB/s        85.528 ms

Asia

Tokyo, Japan            Linode          3.52MB/s        217.539 ms

Singapore               Softlayer       4.31MB/s        165.877 ms

Taiwan                  Hinet           4.00MB/s        271.168 ms

Europe

Rotterdam, Netherlands  id3.net         25.4MB/s        5.205 ms

Haarlem, Netherlands    Leaseweb        20.1MB/s        10.350 ms

(Data supplied by bench.sh 2.0 benchmark script.)

wget Downstream (IPv6)

Code:
Location                Provider        Speed           Latency

United States

Atlanta, GA, US         QuadraNET       603KB/s 99.188 ms

Dallas, TX, US          Linode          678KB/s 114.997 ms

Newark, NJ, US          Linode          12.5MB/s        80.159 ms

Fremont, CA, US         Linode          655KB/s 142.963 ms

Chicago, IL, US         Steadfast       986KB/s 93.033 ms

Asia

Tokyo, Japan            Linode          590KB/s 217.899 ms

Singapore               Linode          602KB/s 161.802 ms

Europe

Frankfurt, Germany      Linode          3.64MB/s        15.175 ms

London, UK              Linode          5.21MB/s        12.843 ms

Haarlem, Netherlands    Leaseweb        8.67MB/s        3.283 ms

(Data supplied by bench.sh 2.0 benchmark script.)

Speedtest.net Speed Test

Code:
  Speedtest by Ookla

     Server: SUPER VPN VIETPN.COM - Amsterdam (id = 19149)

        ISP: HostSlim B.V.

    Latency:     2.55 ms   (0.45 ms jitter)

   Download:   218.20 Mbps (data used: 296.3 MB)

     Upload:   396.61 Mbps (data used: 646.4 MB)

Packet Loss:     0.4%

Result URL: https://www.speedtest.net/result/c/b85d35b3-f456-4e5e-b7f6-b3b85e184018

iperf3 Upstream IPv4 to Serverius (AMS, NL)

Code:
[ ID] Interval           Transfer     Bitrate         Retr

[SUM]   0.00-10.00  sec   579 MBytes   486 Mbits/sec  1509             sender

[SUM]   0.00-10.00  sec   573 MBytes   481 Mbits/sec                  receiver

(Data supplied by the iperf3 tool.)

iperf3 Downstream IPv4 to Serverius (AMS, NL)

Code:
[ ID] Interval           Transfer     Bitrate         Retr

[SUM]   0.00-10.00  sec   245 MBytes   206 Mbits/sec  1180             sender

[SUM]   0.00-10.00  sec   243 MBytes   204 Mbits/sec                  receiver

(Data supplied by the iperf3 tool.)

iperf3 Upstream IPv6 to Serverius (AMS, NL)

Code:
[ ID] Interval           Transfer     Bitrate         Retr

[SUM]   0.00-10.00  sec   606 MBytes   508 Mbits/sec  1335             sender

[SUM]   0.00-10.00  sec   602 MBytes   505 Mbits/sec                  receiver

(Data supplied by the iperf3 tool.)

iperf3 Downstream IPv6 to Serverius (AMS, NL)

Code:
[ ID] Interval           Transfer     Bitrate         Retr

[SUM]   0.00-10.00  sec   123 MBytes   103 Mbits/sec  2795             sender

[SUM]   0.00-10.00  sec   122 MBytes   102 Mbits/sec                  receiver

(Data supplied by the iperf3 tool.)

Concluding from all the test aboves the Downstream (Download) speed seems to be maxing out at around 200 to 250 Mbit/s max which is about 1/4 of the advertised 1 Gbit/s speed. I'm fine with that to be honest but I wonder if there is some kind of throttle in place? The Upstream (Upload) however seems to be maxing out at around 400 to 500 Mbit/s max which is about 2/4 of the advertised 1 Gbit/s speed. Sounds fine to me, too. Although it is a bit weird to see the upload speed being 2x higher than the download speed. Normally if in any case it is otherwise. However usually data center Internet connections are symmetric and so both speeds should be around the same with little to no difference.

However I'm a little disappointed that the IPv6 network performance is rather low. IPv6 is the future and should be adopted as fast as possible. A slow network over IPv6 might however make it not so great in terms of usage. For now I'm fine with this as IPv6 is not so widespread and the biggest amount of traffic is still made over IPv4. Maybe Hostlease doesn't advertise IPv6 for this very reason?

For my purposes the available network speed, especially around the area of Europe, is actually more than enough.

Raw benchmark data is available here: https://mega.nz/#!ZUdwhaYJ!SSczS7esabd1KgXkHLozBKA1i_5yd8qD9liadAsWVUU

Summary & Finale

After all the data gathering and crunching it is time to finalize the review. I will now provide a summary of my experience so far based on the benchmarks and my usage of the VPS so far. And I will add a final opinion about the VPS at the end of this review.

1. The VPS performance is absolutely enough or my VPS usage plans so far. Although there is always more space for better performance. There is probably not much you can do without actually replacing the whole physical server with much newer hardware to improve the performance. So its alright

.

2. The disk performance seems fair enough but a little slow for a SSD setup. Please note again that we don't know the disk setup on the physical server and the load caused by other containers is also unknown. For my purpose it is enough.

3. The network is not really as advertised but I don't really know why. I'm not going to bother Post4VPS staff or Hostlease with this though. I'm fine with the network how it is. However maybe if Hostlease sees this review they can look into a few things?

In my opinion this VPS is a perfect fit if you require a lot of RAM, disk space and several CPU cores. Despite the dated hardware specifications you still get absolutely fair performance. The VPS is absolutely usable for many projects and much more. I would like to thank Post4VPS and Hostlease for the great VPS.

Thank you for reading my review. If you have any questions, suggestions or feedback feel free to post them below.