The EZPZ Hosting Cloud Vs The Big 3

One of the most common questions we get from customers thinking about upgrading to a Cloud VPS is “How fast is it and how does it compare to AWS?”. Normally that’s a very difficult question to answer so we decided to test it out, and a couple of other providers, for ourselves.

Before we dive in to the tests and results there are other factors that should also be taken into consideration including the level of support provided, how fast those support responses are, how flexible the billing and packages are and the performance… so let’s find out.

I’m sure most would assume that companies such as Amazon, Microsoft and Google provide the fastest servers possible and can scale infinity. How do they stack up against the meticulously designed cloud from EZPZ Hosting?

Preparing the VMs

In order to ensure the tests were fair I used the same specs on each cloud, 2 CPU cores, 4GB RAM and SSD storage on CentOS 7 64bit. The servers were also fully updated and rebooted before the tests were started.

On Amazon AWS I used package.c5.large and on Microsoft Azure I used the Standard B2s plan. Google Cloud doesn’t provide a standard package with the same specs so a custom plan was created. All claim to use SSDs and I made sure to change the default Google Cloud boot drive to SSD too.

It was important that we ran tests that simulated real world production usage, not just run a simple drive speed or download test. With this in mind I settled on UnixBench, this will test the hardware performance in a very repeatable way while giving us a score to compare after each test.

I will also be running random read/write tests of various block sizes using fio to test the storage speeds. Using fio is better than a simple dd test in this case as it more closely replicates real world usage, accessing data randomly across the disks in parallel. I will take a total number of IOPS across all 9 tests for a comparison at the end.

Running Unix Bench

Unix Bench is an open-source application available on GitHub for all to see. Installation and usage is simple:

yum -y install gcc gcc-c++ make libXext-devel perl perl-Time-HiRes git
git clone
cd byte-unixbench/UnixBench/

Running fio

Random read/write performance test:

fio --randrepeat=1 --ioengine=libaio --direct=1 --gtod_reduce=0 --name=test --filename=random.fio --bs=4k --iodepth=64 --size=4G --readwrite=randrw --rwmixread=75 | grep IOPS; rm -f random.fio
fio --randrepeat=1 --ioengine=libaio --direct=1 --gtod_reduce=0 --name=test --filename=random.fio --bs=256k --iodepth=64 --size=4G --readwrite=randrw --rwmixread=75 | grep IOPS; rm -f random.fio
fio --randrepeat=1 --ioengine=libaio --direct=1 --gtod_reduce=0 --name=test --filename=random.fio --bs=4m --iodepth=64 --size=4G --readwrite=randrw --rwmixread=75 | grep IOPS; rm -f random.fio

Random read performance test:

fio --name=randread --ioengine=libaio --iodepth=16 --rw=randread --bs=4k --direct=1 --size=4G --numjobs=2 --runtime=240 --filename=random.fio --group_reporting | grep IOPS; rm -f random.fio
fio --name=randread --ioengine=libaio --iodepth=16 --rw=randread --bs=256k --direct=1 --size=4G --numjobs=2 --runtime=240 --filename=random.fio --group_reporting | grep IOPS; rm -f random.fio
fio --name=randread --ioengine=libaio --iodepth=16 --rw=randread --bs=4m --direct=1 --size=4G --numjobs=2 --runtime=240 --filename=random.fio --group_reporting | grep IOPS; rm -f random.fio

Random write performance test

fio --name=randwrite --ioengine=libaio --iodepth=1 --rw=randwrite --bs=4k --direct=1 --size=4G --numjobs=2 --runtime=240 --filename=random.fio --group_reporting | grep IOPS; rm -f random.fio
fio --name=randwrite --ioengine=libaio --iodepth=1 --rw=randwrite --bs=256k --direct=1 --size=4G --numjobs=2 --runtime=240 --filename=random.fio --group_reporting | grep IOPS; rm -f random.fio
fio --name=randwrite --ioengine=libaio --iodepth=1 --rw=randwrite --bs=4m --direct=1 --size=4G --numjobs=2 --runtime=240 --filename=random.fio --group_reporting | grep IOPS; rm -f random.fio

Results and Conclusion

Well, this is embarrassing… for the other guys.

I really wanted to document this to show a cloud designed by highly trained and knowledgeable technicians can keep up with companies that have an almost unlimited budget – but I didn’t expect us to blow them out of the water in all tests.

The results were incredible

The competition is already more expensive than us, but in order to get the same performance you’d need to spend much, much more with the big three than you would with EZPZ Hosting.

We repeated the tests several times over a 24 hour period to make sure they were as fair as possible, just in case one test was run during peak times or a particularly busy period.

Our Cloud VPS packages offer far better value for money and greater performance when compared to other providers. With us you also get:

  • Fully configurable firewall
  • Fast ticket response times
  • 24 hour live chat support
  • Ultra-fast NVMe disk caching
  • Optional automated daily backups with self-managed bare-metal restores.
  • Fully redundant hardware and infrastructure for maximum uptime
  • Easy to use control panel for server reinstalls and power management
  • Simple cPanel, Plesk and CloudLinux management
  • Easy to understand billing


If you’re interested in experiencing our high-performance cloud first hand feel free to contact us, one of our helpful engineers will be happy to answer any questions you may have or build out a custom solution to your requirements.