Using a VPN Doesn't Have To Suck!

FeatherVPN is the FIRST VPN solution that does not slow your network performance.1

A Game Changing VPN

With FeatherVPN there is no compromise on privacy or security. Your traffic is secured using the latest ChaCha20 cipher and TLS1.3 tools. Inspite of these, we have measured network download speeds between 2x and 10x faster than when not using a VPN at all.

2X-10X Speedup Over NO-VPN is Unmatched Performance

In PC Magazine’s list of Best Android VPN’s of 2020, every tested VPN reduced data speeds. The average reduction was 2.5X — which sucks!

LinkRoaming (TM)

An always-on VPN should keep you connected. Today’s devices switch networks all the time. FeatherVPN’s network transitions are seamless.

Carrier Grade Backend

We use BADU Networks' patented WarpEngine-X to provide optimized network transit. With FeatherVPN high latency is no problem.

Measuring Performance is Tricky

Measuring performance is simple to initiate – just run a speed test. Nevertheless, there is a modest number of key factors that can impact speed test results that are often missed, not understood, or simply not known at the time of the test. At the time of the test, a network test on a mobile device is dependent on a combination of factors including and not limited to the behavior of the network, any other activity on the device, data plan details (if it is in use), any quality of service settings between the device and the speed-test server, overall network-latency, network jitter, raw transmission bandwidth, power remaining in the device, wireless bandwidth allocation strategies, load on the speed test server(s), OS version, device driver implementation details, and how the test operates.

Methodology

The preceding paragraph shows that no one, not even the best technologists, can fully explain a particular speed test result. So where does that leave performance measurements, the purpose of which, is to confirm for the end-user that they are using the best tools? To that end, the FeatherVPN team has developed the following test methodology for comparing FeatherVPN against other VPN’s and against the native network.

  1. For each speed test service used to measure performance, complete items 2 to 8.
  2. Define a trial as an equal number of runs of the speed test.
  3. Use the same device for all tests and ensure that the operating system is not using the network prior to the trial run.
  4. Make sure that the test device is fully charged.
  5. Run a NO-VPN baseline trial using mobile, wifi, and host network.
  6. Run a VPN trial using mobile, wifi, and host network.
  7. Run all trials on same mobile network provider.
  8. Run all trials in the same location and if possible against the same test server.
  9. Run all trials sequentially back-to-back.

The idea is to make sure that as many as possible of the factors listed earlier that can impact a speed test result are the same for all trials, thereby, the ensuing test results can be relied upon. It is possible that tests conducted in a different location, or using a different devices, or different host network providers, may differ. When such an anomaly from predicted results is detected, a sound methodology makes it easier to investigate its cause.

Fast.com and Speedtest.net Results


We ran the above tests using a number of Android devices the results shown are from a Samsung Galaxy S20, using both WiFi and an unlimited T-Mobile USA LTE data plan. All tests were run in the same location at Le Pain Quotidien in Fashion Island, Newport Beach, CA on the T-Mobile 5G network. The results shown are indicative of behavior of FeatherVPN using the speedtest.net test apps. The behavior was consistent across all trial runs although speedtest.net produces higher test run variance than Fast.com. In our opinion Fast.com is a more consistent and accurate network performance test.

Carrier Grade Backend

The most difficult, and important, part of creating a VPN service is selecting the backend. Almost anyone can build an App these days. However, a VPN service requires two key components; encryption/decryption, and carrier grade scalability for the VPN servers. The former is provided by using appropriately licensed open source software, but the latter requires an efficient backend solution.

To build the backend many VPN providers rely on cloud hosted virtual instances of open source and free OpenVPN server (or similar). While free to deploy, this solution is anything but high performance.

Instead, the FeatherVPN engineers chose to deploy an exclusively customized BADU Networks' WarpEngine-X. Network session optimization is the key reason that FeatherVPN has such great performance.


WarpEngine-X scales to millions of sessions per server and provides sub-millisecond session optimization. It has proved to be highly reliable with five 9’s reliability. WarpEngine-X is used by BADU Networks' enterprise and telecom customers to provide network optimization for both short and long round trip times and in MVNO environments.

As shown, the optimization routinely yields for different content sizes network transit improvements of between 2X and 10X.

In summary, every part of FeatherVPN is engineered for performance, from be the backend servers to the low power consumption clients. The resulting solution provides a sound platform for delivering a high-speed VPN with LinkRoaming and best of class security.