Never a dull moment in the world of VPS providers. In the past month,
DigitalOcean switched up their C-suite and Linode opened up a data center in
Mumbai, India.
I’m switching things up a bit as well, bringing things back to the east coast.
Well mostly, since not everybody has a New York-based data center.
I’ve also left the slightly more expensive high frequency instance on the
comparison. Since it performed so well last month for a trivial amount more per
month, I was curious if it would stack up again this month.
As always, I’ve built out three servers with each provider, running Ubuntu 18.04
LTS. The results are an average of the results from each of the instances.
Also, I usually don’t talk much about the support from the different providers,
since I find it hard to truly test, but I did get some insight from my buddy
Justin (from ReviewCycle) about his recent DigitalOcean support
interaction.
He happened to have borked a configuration file that locked him out of his box
(couldn’t SSH in directly, only via their console) and DigitalOcean’s support
sluffed it off as him needing to enter his password in when prompted (which he
wasn’t).
Overview
| DigitalOcean | Lightsail | Linode | UpCloud | Vultr | |
|---|---|---|---|---|---|
| Location | New York City 1 | Virginia, Zone A | Newark, NJ | Chicago #1 | New Jersey |
| RAM | 1 GB | 1 GB | 1 GB | 1 GB | 1 GB |
| CPU | 1 Core | 1 Core | 1 Core | 1 Core | 1 Core |
| Storage | 25 GB SSD | 40 GB SSD | 25 GB SSD | 25 GB SSD | 32 GB NVMe |
| Transfer | 1 TB | 2 TB | 1 TB | 1 TB | 1 TB |
| Base Price | $5/month | $5/month | $5/month | $5/month | $6/month |
| Backups | $1/month | N/A | $2/month | $0.06/GB | $1.20/month |
| Transfer Overage | $0.01/GB | $0.09/GB | $0.02/GB | $0.056/GB | $0.01/GB |
| Load Balancer | $10/month | $18/month | $10/month | N/A | N/A |
CPU Info
| DigitalOcean | Lightsail | Linode | UpCloud | Vultr | |
|---|---|---|---|---|---|
| CPU MHz | 2200.00 | 2400.22 | 2300.00 | 2999.99 | 3792.00 |
| Cache Size (KB) | 30720.00 | 30720.00 | 16384.00 | 16384.00 | 16384.00 |
| BogoMips | 2933.00 | 3200.33 | 3066.33 | 3999.67 | 5056.33 |
CPU
| DigitalOcean | Lightsail | Linode | UpCloud | Vultr | |
|---|---|---|---|---|---|
| Events per Second | 744.64 | 863.31 | 746.46 | 1073.58 | 1311.61 |
| Minimum (ms) | 1.19 | 1.03 | 1.28 | 0.91 | 0.74 |
| Average (ms) | 1.38 | 1.16 | 1.34 | 0.93 | 0.76 |
| Maximum (ms) | 9.92 | 1.37 | 20.97 | 2.65 | 3.42 |
Memory (Read)
| DigitalOcean | Lightsail | Linode | UpCloud | Vultr | |
|---|---|---|---|---|---|
| Ops per Second | 2995704.71 | 824115.01 | 3287071.06 | 4518478.03 | 5352398.30 |
| Minimum (ms) | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| Average (ms) | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| Maximum (ms) | 7.65 | 1.60 | 3.26 | 2.13 | 0.71 |
Memory (Write)
| DigitalOcean | Lightsail | Linode | UpCloud | Vultr | |
|---|---|---|---|---|---|
| Ops per Second | 2992589.12 | 824752.56 | 3268418.59 | 4571707.28 | 5380630.02 |
| Minimum (ms) | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| Average (ms) | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| Maximum (ms) | 5.59 | 1.54 | 9.47 | 0.88 | 3.78 |
File I/O
| DigitalOcean | Lightsail | Linode | UpCloud | Vultr | |
|---|---|---|---|---|---|
| Reads per Second | 1539.46 | 1243.71 | 1847.02 | 3511.00 | 4928.12 |
| Writes per Second | 1026.27 | 829.14 | 1231.34 | 2340.66 | 3285.35 |
| Fsyncs per Second | 3275.67 | 2649.21 | 3933.88 | 7482.26 | 10506.72 |
| Minimum (ms) | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| Average (ms) | 0.20 | 0.22 | 0.15 | 0.08 | 0.05 |
| Maximum (ms) | 92.92 | 14.63 | 15.56 | 35.51 | 3.88 |
MySQL
| DigitalOcean | Lightsail | Linode | UpCloud | Vultr | |
|---|---|---|---|---|---|
| Transactions per Second | 2559.00 | 2238.33 | 2817.33 | 4507.67 | 6430.67 |
| Queries per Second | 51180.00 | 44766.67 | 56346.67 | 90153.33 | 128613.33 |
| Minimum (ms) | 2.98 | 2.33 | 2.11 | 1.56 | 1.19 |
| Average (ms) | 5.45 | 4.47 | 3.56 | 2.22 | 1.56 |
| Maximum (ms) | 56.62 | 70.37 | 31.19 | 24.37 | 15.91 |
Redis
| DigitalOcean | Lightsail | Linode | UpCloud | Vultr | |
|---|---|---|---|---|---|
| PING_INLINE | 30778.03 | 55916.05 | 38132.83 | 67300.54 | 65023.11 |
| PING_BULK | 29519.50 | 55383.74 | 39511.71 | 66805.65 | 64632.57 |
| SET | 30868.78 | 55466.07 | 39957.22 | 67309.43 | 65288.79 |
| GET | 29596.39 | 55524.81 | 40043.16 | 66127.06 | 65249.19 |
| INCR | 29886.55 | 56362.21 | 40195.95 | 67486.28 | 65708.38 |
| LPUSH | 31325.69 | 55252.36 | 39818.74 | 68712.79 | 65577.45 |
| RPUSH | 31693.48 | 56043.15 | 40074.20 | 68239.07 | 66001.58 |
| LPOP | 31119.83 | 55163.36 | 38769.57 | 68033.69 | 66299.75 |
| RPOP | 29984.15 | 55401.72 | 39325.47 | 67720.17 | 66046.04 |
| SADD | 31139.83 | 56403.44 | 39591.98 | 67657.89 | 65690.84 |
| HSET | 30775.99 | 54170.39 | 40246.81 | 68034.42 | 65648.46 |
| SPOP | 30273.19 | 56107.30 | 39021.95 | 67123.56 | 65410.08 |
| LRANGE_100 (first 100 elements) | 15551.30 | 30063.38 | 19990.79 | 35386.74 | 39119.30 |
| LRANGE_300 (first 300 elements) | 6697.52 | 11986.61 | 8296.46 | 14060.53 | 17259.02 |
| LRANGE_500 (first 500 elements) | 4823.05 | 8431.73 | 5610.11 | 9949.41 | 12454.93 |
| LRANGE_600 (first 600 elements) | 3847.19 | 6696.60 | 4546.09 | 8090.43 | 9956.44 |
| MSET (10 keys) | 25943.18 | 40303.53 | 35170.02 | 53311.16 | 55964.83 |
Speed Test
| DigitalOcean | Lightsail | Linode | UpCloud | Vultr | |
|---|---|---|---|---|---|
| Distance (km) | 2435.00 | 2090.30 | 2397.72 | 1397.78 | 2390.98 |
| Latency (ms) | 44.804 | 46.604 | 47.978 | 36.116 | 47.173 |
| Download (Mbit/s) | 1341.31 | 53.23 | 153.48 | 132.78 | 1465.67 |
| Upload (Mbit/s) | 500.81 | 98.78 | 381.73 | 454.23 | 541.41 |
Apache Benchmark (against nginx on the servers)
| DigitalOcean | Lightsail | Linode | UpCloud | Vultr | |
|---|---|---|---|---|---|
| Requests per Second | 157.73 | 170.25 | 186.28 | 171.76 | 171.76 |
| Time per Request (ms) (mean) | 3177.86 | 2941.56 | 2726.56 | 2935.01 | 2937.44 |
| Transfer Rate (Kbyte/sec) | 132.30 | 142.82 | 156.26 | 144.08 | 144.08 |
Conclusion
Lightsail still stacks up best if you’re looking for the raw product offering,
but Vultr’s high frequency instance stomped the competition in nearly every
category.
The big exception being Redis, where UpCloud stole the show for the majority of
the operations that were run. UpCloud also had some of the lowest latency during
the speed tests.
Not to be forgotten, DigitalOcean put up a great showing overall and Linode put
up some great numbers during this month’s Apache Benchmarks.
As always, you should weigh our your personal needs and do your own independent
research and benchmarks when choosing your hosting provider.
If you happened to have found these benchmarks helpful in your own quest to find
your perfect VPS hosting provider and have decided to sign up, please please
please do so with one of my referral links. It’ll be greatly appreciated!
Also, comment below if you do, I love hearing about who everybody has picked.
