What determines VPN performance?

[Larry Gelencser]

Greetings Guru's,

   What actually determines VPN, particularly lan-to-lan, performance/speed? Is it physical bandwidth on each side or is there specific settings on the devices somewhere?

  The reason I ask is because I'm about to sign a new internet contract for my corp office where I host services such as Email/AV/Database etc for my remote offices. I can increase my speed to either 30Mbps or 50Mbps and the pricing difference is not much so if it'll help my vpn tunnels I'm all for it but perhaps it's null?

Here's my setup:

• Corp Office:  ASA5510 with 20/20Mbps internet
• 17 Remote Office's: RV325 with 15/3Mbps internet

The internet provider at most all my remote offices is on the same provider backbone and I have decent response time between them (avg 30ms). I used to have 7Mbps over copper lines before switching to Metro Ethernet at my corp office and the difference between that and the 20Mbps was substantial but with 2 different technologies I don't know if I'll actually see a performance difference between 30Mbps and 50Mbps but would appreciate your thoughts.

Thanks in advance!

[Karsten Iwen · Accepted solution]

Unless the devices are not the limiting factor (the ASA 5510 isn't for 20/20) it's mainly the bandwidth of the provider-connection that limits the VPN throughput.

More bandwidth is nearly always a good idea, but if the actual bandwidth is not at the limit, then there will be no big benefit.

I would first install a monitoring solution (like Paessler PRTG) and monitor the usage of the VPNs and the external interface of the ASA. Then you see how likely it is that the added external bandwidth will help you.

[Larry Gelencser]

Thanks again Karsten!

   Although I've given up trying to get the Psessler software running as I can't even get past the bogus license portion as it'll never recognize the trial key......

[Karsten Iwen]

They have a good support, it's worth asking them for help to get it running.