It’s about that time again!
Even though the last week has been pretty hectic with travel, I’m still trying
hard to make sure these VPS comparisons are showing up on the first Monday of
the month (which lines up with my usual blogging schedule).
Seems the 5 buck pricing tier is still pretty popular. Nobody has been beating
on my door asking for comparisons of different pricing tiers, so I’m going to
just continue last month’s trend of reviewing the lowest price point that each
of the current providers offer.
As always, the following benchmarks are the average across three VPS servers
with each of the providers. Each box is running Ubuntu 18.04 LTS and the
locations chosen were based on their proximity to the New York area which most
of the providers offer.
Overview
| DigitalOcean | Lightsail | Linode | Vultr | |
|---|---|---|---|---|
| Location | New York 1 | Virginia, Zone A | Newark, NJ | New York (NJ) |
| RAM | 1 GB | 1 GB | 1 GB | 1 GB |
| CPU | 1 Core | 1 Core | 1 Core | 1 Core |
| SSD | 25 GB | 40 GB | 25 GB | 25 GB |
| Transfer | 1 TB | 2 TB | 1 TB | 1 TB |
CPU Info
| DigitalOcean | Lightsail | Linode | Vultr | |
|---|---|---|---|---|
| CPU MHz | 2294.61 | 2399.95 | 2433.33 | 2398.15 |
| Cache Size (KB) | 25344.00 | 30720.00 | 16384.00 | 16384.00 |
| BogoMips | 3059.67 | 3200.33 | 3333.00 | 3196.00 |
CPU
| DigitalOcean | Lightsail | Linode | Vultr | |
|---|---|---|---|---|
| Events per Second | 794.19 | 863.55 | 750.35 | 820.68 |
| Minimum (ms) | 0.98 | 1.07 | 1.25 | 1.13 |
| Average (ms) | 1.29 | 1.16 | 1.33 | 1.22 |
| Maximum (ms) | 3.56 | 3.72 | 2.27 | 7.15 |
Memory (Read)
| DigitalOcean | Lightsail | Linode | Vultr | |
|---|---|---|---|---|
| Ops per Second | 3503034.90 | 824876.36 | 1553589.88 | 3450400.74 |
| Minimum (ms) | 0.00 | 0.00 | 0.00 | 0.00 |
| Average (ms) | 0.00 | 0.00 | 0.00 | 0.00 |
| Maximum (ms) | 0.30 | 0.20 | 0.76 | 2.85 |
Memory (Write)
| DigitalOcean | Lightsail | Linode | Vultr | |
|---|---|---|---|---|
| Ops per Second | 3619458.94 | 824663.80 | 1557095.18 | 3349987.46 |
| Minimum (ms) | 0.00 | 0.00 | 0.00 | 0.00 |
| Average (ms) | 0.00 | 0.00 | 0.00 | 0.00 |
| Maximum (ms) | 0.88 | 2.03 | 1.15 | 6.45 |
File I/O
| DigitalOcean | Lightsail | Linode | Vultr | |
|---|---|---|---|---|
| Reads per Second | 959.77 | 1205.07 | 1587.86 | 1721.03 |
| Writes per Second | 639.81 | 803.37 | 1058.57 | 1147.32 |
| Fsyncs per Second | 2039.77 | 2567.17 | 3379.15 | 3668.12 |
| Minimum (ms) | 0.00 | 0.00 | 0.00 | 0.00 |
| Average (ms) | 0.33 | 0.22 | 0.17 | 0.15 |
| Maximum (ms) | 14.15 | 51.79 | 21.63 | 64.27 |
MySQL
| DigitalOcean | Lightsail | Linode | Vultr | |
|---|---|---|---|---|
| Transactions per Second | 1619.33 | 2065.00 | 2331.33 | 2522.33 |
| Queries per Second | 32386.67 | 41300.00 | 46626.67 | 50446.67 |
| Minimum (ms) | 3.34 | 2.93 | 2.34 | 2.22 |
| Average (ms) | 6.88 | 4.87 | 4.62 | 3.99 |
| Maximum (ms) | 51.69 | 48.84 | 44.60 | 45.82 |
Speed Test
| DigitalOcean | Lightsail | Linode | Vultr | |
|---|---|---|---|---|
| Distance (km) | 2434.95 | 2090.30 | 2421.10 | 2391.13 |
| Latency (ms) | 48.824 | 43.630 | 50.615 | 46.368 |
| Download (Mbit/s) | 1315.66 | 204.70 | 252.16 | 290.51 |
| Upload (Mbit/s) | 512.84 | 302.04 | 316.96 | 543.03 |
Apache Benchmark (against nginx on the servers)
| DigitalOcean | Lightsail | Linode | Vultr | |
|---|---|---|---|---|
| Requests per Second | 263.23 | 264.89 | 251.15 | 220.30 |
| Time per Request (ms) (mean) | 1900.03 | 1888.72 | 2009.03 | 2303.17 |
| Transfer Rate (Kbytes/sec) | 220.82 | 222.21 | 210.68 | 184.80 |
Conclusion
As always, these benchmarks should be analyzed based on your own needs and not
necessarily taken at face value.
If we are to talk about them at face value, Lightsail excels in regard to the
storage and bandwidth offering compared to the rest of the pack.
The specifications of the CPU were fairly consistent across the board with
Lightsail processing the most requests per second.
For memory, both reads and writes, DigitalOcean crushed it with Vultr not far
behind. Worth noting, even with Vultr putting up great numbers, their max
execution times were fairly abysmal compared to the rest.
Vultr’s file I/O was ahead but consistent with the memory benchmarks, the
maximum execution time was was up there. The MySQL benchmarks also saw Vultr
ahead of the pack for the number of queries and just squeaking out on
transactions.
The speed test was dominated by DigitalOcean for downloads and coming in second
against Vultr for uploads. The ever controversial ab benchmark was Lightsail
pull ahead with Linode not far behind.
If you found these benchmarks helpful and you’re planning on pulling the trigger
on a new server soon, please take a moment to do so via one of my referral
links.
