What Makes A High Speed VPN Service Provider & How to Speed Test a VPN

What Makes a High Speed VPN Service Provider

At VikingVPN, we strive to be the fastest and most reliable VPN service out there. We want to be a service you can always leave on without worrying about the performance impact of the encryption, or the reliability of the VPN servers. This page is about what makes a VPN fast, and what we do to set ourselves apart from the pack.

VikingVPN maintains the speediest VPN network in the world.  Here's a sample speed test run from a Google Fiber 1 Gigabit connection in Kansas City to our Chicago Server Cluster:

Things that impact the performance of your VPN service:

1. The quality of the network.

This is the connections that the servers have to the internet, and the quality of their connections to you, and the places you want to visit on the net. VikingVPN is a premium VPN service and spends enormous resources making sure we have access to the fastest hosts possible with many parallel high quality links to the internet. This means we enjoy phenomenal uptime (99.999% is our goal, which allows for 15 minutes of downtime per year, and we have exceeded that goal since exiting beta). We also enjoy fantastic overall speeds for our customers.

2. The density of users on the network.

This is the number of connections that are active per server. Servers have limited resources, and it is important to understand the practices of overselling. If you have too many users connecting to a single server, it doesn't matter how great the datacenter is. The service is going to suffer because the resources are being overutilized. A good analogy is it doesn't matter whether your street is a country road or a 10-lane interstate highway, if you have too many cars on the road traffic jams are going to occur. VikingVPN has policies specifically addressing user density that require us to expand our network when a user density threshold is reached.

3. Your distance from the VPN server.

Because of how anonymous VPN services work, you want to be as close to your server as possible. The further that you are from your server, the more performance degrades. A good analogy for this is trying to pull a heavy object with a short rope, and then trying to pull that same object with a rope a half mile long.

4. The limits of your devices.

It is important to understand that you cannot exceed the physical limitations of your home network. If you are using your VPN over a 802.11g wireless connection, you have in a best case scenario 54Mbps, and more than likely a lot less than that due to signal degradation, wireless congestion, and wireless collisions. As you get into the newer standards like 802.11n, 802.11n+, and 802.11ac, you get higher theoretical throughput, but these signals are in fact WEAKER than their older counterparts because they require wider wireless channels or higher frequencies that do not travel as well as their lower frequency cousins. You are far better off using a wired internet connection to avoid these issues entirely where possible, especially if you are trying to get very high speeds.

How to properly speed test a VPN

It is very important to note that using flash-based speed tests like speedtest.net are unrealible for VPN services. This is because VPN services use compression to increase the speed that data flows across the network. Some flash based speed test servers do not properly account for this. Your best bet is to use real world on/off testing with legal torrents, or test files from web hosting services. These tend to give you very accurate real world results.

A naked Speedtest.net run of a RCN 110mbit (13.75MB) connection.

An inaccurate speed test of that same connection on VikingVPN. This was taken seconds later on the same Speedtest.net server just signed on to the VPN. (15.80 MB/sec or 126.37 Mbit/sec) No, the download of the connection did not increase in bandwidth by 50% due to being on a VPN. This is just a false read due to compression. It is easy to tell in this case because it far exceeds the speeds offered by the ISP.

A test file download from a local Chicago web host shows an accurate read. (7.6 MB/sec or 60.8 Mbit/sec)

A legal torrent download also shows a more accurate read. Speed with this connection would actually probably be higher, but I had trouble finding a download that was both large enough and had enough seeds for this demonstration. (6.5 MB/sec or 52 Mbit/sec)

Services that we recommend for speed testing VikingVPN:

Test files to download:

If you are connected to Chicago: http://dft.singlehop.com/100megabytefile.tar.gz

If you are connected to New York City: http://testfile.nyc.steadfast.net/data.bin

If you are connected to Amsterdam:

If you are connected to Seattle:

For Bittorrent:

Make sure you follow our torrent optimization guide for optimal speeds.

Then select a large legal torrent with a lot of seeds and give it a couple of minutes to get going and connected to a lot of seeds and peers so you can see the real speeds. That should give you a pretty accurate read of real world speeds!

Speed test websites that account for compression properly for VPNs:

Update April 4th 2015: Both of these sites have been unreliable lately for connections over 20 Mbit.