Comparing Apple M1 Performance to VPS Servers

Last updated on

On November 17, 2020 Apple released the MacBook Air with the M1 chip. With all the hype surrounding the release I thought VPSBenchmarks need not be left behind and could add its own benchmarks to provide a server perspective on this chip.

The comparison to the same benchmarks results for datacenter servers should be considered a curiosity more than an apple to apple comparison. For starters the M1 SoC is very different from a traditional CPU and it's not designed to be used on shared datacenter servers. In addition, the tested instance was the sole tenant of a laptop and was able to use 100% of the resource that it was allocated.

The Setup

  • A new MacBook Air with the M1 chip, 8GB RAM and 250GB storage.
  • The Parallels Desktop Technical Preview for M1 chip.
  • The VBSBenchmarks test suite was installed inside an Ubuntu 20.04.1 aarch64 VM running inside Parallels and was allocated 2 CPU, 20GB disk and 2GB RAM.

This lscpu output on the Ubuntu VM does not let a lot of information transpire about the CPU, there's no frequency and there's not even a model name:

$ lscpu
Architecture:                    aarch64
CPU op-mode(s):                  64-bit
Byte Order:                      Little Endian
CPU(s):                          2
On-line CPU(s) list:             0,1
Thread(s) per core:              1
Core(s) per socket:              2
Socket(s):                       1
NUMA node(s):                    1
Vendor ID:                       ARM
Model:                           0
Stepping:                        r0p0
BogoMIPS:                        48.00
NUMA node0 CPU(s):               0,1
Vulnerability Itlb multihit:     Not affected
Vulnerability L1tf:              Not affected
Vulnerability Mds:               Not affected
Vulnerability Meltdown:          Not affected
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:        Mitigation; __user pointer sanitization
Vulnerability Spectre v2:        Not affected
Vulnerability Srbds:             Not affected
Vulnerability Tsx async abort:   Not affected
Flags:                           fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt
                                  fcma lrcpc dcpop sha3 asimddp sha512 asimdfhm dit uscat ilrcpc flagm ssbs sb paca p
                                 acg dcpodp flagm2 frint

The Environment

  • The laptop was plugged in for the entire duration of the tests.
  • All desktop software on the Mac susceptible of using CPU or disk was turned off.
  • The laptop was connected to wifi on my home network connection making the Network Transfer tests irrelevant. They crashed anyway with a segfault in libcurl, I blame the Parallels Tech Preview for that.

The Tests

A VPSBenchmarks trial normally involves Web, Sysbench, Network and Endurance tests. In this case, the network tests are irrelevant as explained above.

The Endurance test main goal is to measure performance stability and variations in availability of resources. For a single VM instance on an idle laptop, performance stability is a given so the Endurance tests were not run.

We're left with the Web Runs and Sysbench tests which are perfect to test CPU and storage performance. Anecdotally the MacBook Air got perceptively warm but not hot during the web runs at high request rates. There's no fan in the MacBook Air M1 so the laptop stayed silent at all times.

The Results

The M1 didn't disappoint, the CPU benchmarks demonstrate it's in a class of its own. All the results can be reviewed from the Apple M1 trial page.

Since all test results are needed to calculate the grades and this trial only has Web Runs and Sysbench, this trial has no grade.

Web Tests

Here is a comparison of the Apple M1 web tests results to other 2 cores instances running on Xeons or EPYCs.

The HTTP latency of the M1 is half the latency of the fastest VPS in this sample.

In terms of traffic volume that the instance can handle, the M1 2 cores was able to handle 80 requests per second without error or timeout during our web tests. That's 30 requests per second faster than the next fastest 2 cores VPS in the sample.

Sysbench CPU Test

Here is a Sysbench comparison with the same servers sample as above.

The chart says it all:
Sysbench CPU

Sysbench Disk IO

The disk tests are not as meaningful as the CPU tests, the M1 plays no part in the speed of the SSD inside the laptop. I expect the results would be the same for a same generation MacBook Air with x86 inside.

See for yourself, there are a number of disk IO tests in the comparison page. The MacBook comes on top for some of them but it's just a bit better than average on others.


There have been plenty of articles explaining why the M1 is so much faster than current Intel or AMD desktop CPUs. See for example the detailed explanations in Erik Engheim's blog post. Regarding the server benchmarks above, it comes down to:

  • 5nm manufacturing process.
  • Ability to pack more instruction decoders in RISC processors as opposed to CISC.
  • Physical proximity of the memory to the actual processing units.

On thing is certain, with such an innovative SoC design, Apple is going to turn the CPU world upside down. We're hopefully going to see more innovation soon from other chip designers and this upheaval will benefit the cloud servers industry in addition to the consumer market. Cloud providers need open, affordable and performant SoC systems. Intel and AMD have their work cut out for them.


Find the VPS that fits your requirements in seconds


Best VPS 2022

Provider Finder

Run a fast benchmark suite with Geekbench, Fio and iPerf3 on your server, visualize and compare the results for free:
Get Started

Share this page

Latest Tweets by @vpsbenchmarks

Provider News and Promotions

Vultr Managed Databases features and prices In July 2022, Vultr released Vultr Managed ...
Multi-datacenter clustering at ScalaHosting In May 2022, ScalaHosting released its new cluster management ...



Sysbench fileio read tests and the --validate option Late in March 2022 a provider reported to VPSBenchmarks a flaw in the sysbench fileio read metrics: ...


Be Featured on VPSBenchmarks

Cloud Server Providers, would you like your products to be tested at

Check out our pricing page and sign up


We are proud to provide objective and impartial benchmark data on this website. VPSBenchmarks receives support from some of the providers featured here but all tests are conducted the same way regardless of our relationships with them.

If you find our benchmarks valuable, you can help by making your cloud purchases using the Provider Affiliate Buttons displayed throughout the site.

Be the first to learn about new Best VPS rankings. Subscribe to our newsletter.