VPN Speedtest.net Throughput

[RRRRRR]

I’m connecting to a VPN server on another residential ISP about 200mi away (cable) from my location (Fiber). My connection speeds via speedtest.net are as follows..

My Local: 153MB down / 74MB up (Fios), 34ms ping
VPN Server (tested directly from machine): 62MB down / 10MB up (cable), 12ms ping
Me connected to VPN Server: 0.49MB down / 3.8MB up, 90ms ping

My machine is 2012 Macbook Pro connected via ethernet and gigabit router, OSX 10.9. Server is also connected via ethernet behind residential gateway with 1723 opened, Mac Mini with OSX 10.8. VPN is connected via PPTP using iVPN and Mac OS’s built in VPN.

My question is simple, why do I see 0.5MB speeds through this pipe when everything on both ends tests perfectly. Shouldn’t I be seeing something near the full upload pipe of the VPN server (10MB/10MB). This seems to happen every day in the evening hours more than non peak hours... but again both sides test flawlessly for their full purchased speeds at speedtest.net locally at all times.

Is this just a matter of too many hops from residential to residential connections where I’m losing packets or something? I have been trying to figure this out for months on my own with no success.

 

[bill001g]

You would need to test from the other location to the server without the VPN to see if there was some issue with the latency. You will need to use a tool like IPERF or something and you cannot directly compare the numbers to speed test sites. You should if you work at it be able to run iperf both though the vpn and directly so you can compare.

VPN has a lot of overhead in it. Not as much as you show though. You also have to worry about cpu load on both devices.



I suspect your

 

[RRRRRR]

You would need to test from the other location to the server without the VPN to see if there was some issue with the latency. You will need to use a tool like IPERF or something and you cannot directly compare the numbers to speed test sites. You should if you work at it be able to run iperf both though the vpn and directly so you can compare.

VPN has a lot of overhead in it. Not as much as you show though. You also have to worry about cpu load on both devices.




I suspect your

Your suspicion was cut off and I wanted to hear it.

From iperf with vpn not connected...

Client connecting to [serveripprivate], TCP port 5001
TCP window size: 1.00 MByte (WARNING: requested 1.00 MByte)
------------------------------------------------------------
[ 4] local 192.168.1.211 port 64602 connected with [serveripprivate] port 5001
[ ID] Interval Transfer Bandwidth
[ 4] 0.0- 1.0 sec 1.62 MBytes 1.62 MBytes/sec
[ 4] 1.0- 2.0 sec 1.00 MBytes 1.00 MBytes/sec
[ 4] 2.0- 3.0 sec 1.38 MBytes 1.38 MBytes/sec
[ 4] 3.0- 4.0 sec 1.50 MBytes 1.50 MBytes/sec
[ 4] 4.0- 5.0 sec 1.75 MBytes 1.75 MBytes/sec
[ 4] 5.0- 6.0 sec 2.00 MBytes 2.00 MBytes/sec
[ 4] 6.0- 7.0 sec 2.12 MBytes 2.12 MBytes/sec
[ 4] 7.0- 8.0 sec 2.38 MBytes 2.38 MBytes/sec
[ 4] 8.0- 9.0 sec 2.75 MBytes 2.75 MBytes/sec
[ 4] 9.0-10.0 sec 2.88 MBytes 2.88 MBytes/sec
[ 4] 10.0-11.0 sec 3.12 MBytes 3.12 MBytes/sec
[ 4] 11.0-12.0 sec 3.00 MBytes 3.00 MBytes/sec
[ 4] 12.0-13.0 sec 1.88 MBytes 1.88 MBytes/sec
[ 4] 13.0-14.0 sec 2.00 MBytes 2.00 MBytes/sec
[ 4] 14.0-15.0 sec 2.12 MBytes 2.12 MBytes/sec
[ 4] 15.0-16.0 sec 1.25 MBytes 1.25 MBytes/sec
[ 4] 16.0-17.0 sec 1.50 MBytes 1.50 MBytes/sec
[ 4] 17.0-18.0 sec 1.62 MBytes 1.62 MBytes/sec
[ 4] 18.0-19.0 sec 1.88 MBytes 1.88 MBytes/sec
[ 4] 19.0-20.0 sec 2.12 MBytes 2.12 MBytes/sec
[ 4] 20.0-21.0 sec 2.25 MBytes 2.25 MBytes/sec
[ 4] 21.0-22.0 sec 2.50 MBytes 2.50 MBytes/sec
[ 4] 22.0-23.0 sec 1.75 MBytes 1.75 MBytes/sec
[ 4] 23.0-24.0 sec 1.50 MBytes 1.50 MBytes/sec
[ 4] 24.0-25.0 sec 1.88 MBytes 1.88 MBytes/sec
[ 4] 25.0-26.0 sec 2.12 MBytes 2.12 MBytes/sec
[ 4] 26.0-27.0 sec 2.12 MBytes 2.12 MBytes/sec
[ 4] 27.0-28.0 sec 2.50 MBytes 2.50 MBytes/sec
[ 4] 28.0-29.0 sec 2.75 MBytes 2.75 MBytes/sec
[ 4] 29.0-30.0 sec 2.75 MBytes 2.75 MBytes/sec
[ 4] 0.0-30.0 sec 62.1 MBytes 2.07 MBytes/sec


Ok so it looks as if the VPN protocol isn't the one to blame right? If that's the case, why is there such terrible transfer throughput speeds between these machines?

 

[bill001g]

It appears you do not have a generic network issue between the locations. You want to use the option to run the test in both directions though to test up and down.

This means (and what I was going to put as a suspect) it is likely some software issue with the VPN server. It can either be the encryption process itself or the issue with the routing of the traffic. You should be able to use IPERF to test the encryption. You will want to run iperf to a destination address on the lan on the server side. You might be able to use the lan address of the VPN server but sometime that confuses the server. It tends to be easier to run to another PC on the lan. If things are working correctly iperf should think it is talking between 2 lan devices.

If this is working you should see about a 10% or so loss in speed due to the VPN overhead compared to your results without the vpn up.

This would leave the routing issue. Normally the symptom of this problem is that it just does not work at all but the vpn box and the router may be getting confused. You have 2 streams of data. The first is from you to the VPN box over the tunnel. The router will be sending that data to the lan address of the vpn box. Now the vpn box will strip off the encryption and now your real traffic with a lan address will leave the vpn box on the same port it arrived and be sent back to the router. Since you now have 2 ips and 2 mac addresses coming from the same physical device things can get messed up pretty easy. Still this usually causes issues where it just doesn't work slowdowns would imply some kind of packet loss.

 

[djin45]

My question is simple, why do I see 0.5MB speeds through this pipe when everything on both ends tests perfectly. Shouldn’t I be seeing something near the full upload pipe of the VPN server (10MB/10MB). This seems to happen every day in the evening hours more than non peak hours... but again both sides test flawlessly for their full purchased speeds at speedtest.net locally at all times.

Is this just a matter of too many hops from residential to residential connections where I’m losing packets or something? I have been trying to figure this out for months on my own with no success.

I'd advise you to use some enother VPN service. Here is one I use myself www.faceless.me
Besides they heve a couple interesting features:
- It's safe!
- It has FREE tariff, but with some ad`s
- Hide your IP
- It's contains protocols from OpenVPN and PPTP
- It`s multilanguage
- VPN will not save your logs!