Layer 2 VPN vs Layer 3 VPN

Good morning friends,

The below quote is from user “TK_Liew”

“Choosing L2 or L3 SpeedFusion is based on your need. If multicast is needed between each location then L2 SpeedFusion is recommended. Else L3 will be the choice because will have better management and control.”

I am looking to deploy a global solution that requires multi-casting and it seems that Layer 2 VPN is my only choice.I am concerned to what are the Layer 2 VPN limitations from a management and control point of view? Also is a Layer 2 VPN as reliable as Layer 3 VPN?

Enjoy your weekend.


After the Layer 2 SpeedFusion was established, both of the Peplink devices will act as Layer 2 switches. Communication for Layer 3 and above will control by other devices. This is what I mean.

Layer 2 or 3 having the same reliability. Since multicast is needed in your deployment, Layer 2 SpeedFusion will be more suitable.

1 Like

Good morning TK_Liew

I was planning to put a Balance 710 in the HQ and then Balance 380’s for the regional office/various counties (about 7); furthermore remote in country users will be connecting via MAx BR1 routers (3G + Satellite) to the regional, in country, Balance 380 offices. The multicasting is required for voice PTT

Based on your comments, who will handle Layer 3 routing? Do you recommend installing an additional Peplink in the HQ?
Will this be a scenario of Star/Mesh Topology?

Thank u

In a L2 Speedfusion setup, Peplink devices only act as the bridge between your offices. The one handling L3 routings are usually the core switch sitting behind your Peplink devices.

An additional Peplink in HQ in a HA setup will be nice, considering your traffic are all routing via HQ.

This should be in a star-topology.
Balance 710 will be connected to the Balance 380s.
Balance 380s will be connected to their respective BR1s.

1 Like

Dear Kv_Chen,

Thank you for this; does Peplink offer design assistance? Since this is a PTT solution and I am trying to have equal latency (as much as I can) between the local field Max BR1, the the local country office and the HQ office, does it make sense for the setup between Balance 710, Balance 380 and Max BR1 to be in a MESH topology while the Balance 710 to all regional/local offices be in a STAR topology?

I am trying to have voice from the field unit to land to two places, local office and HQ.


What I am trying to achieve is a Global Multicasting network that multicasting packets, from ROIP, will be delivered to the local office and also the HQ for redundancy…we are looking in the long run to support approximately 15 countries and 1000 radios. The Speedfusion Hot Fail-over will allow to seamlessly operate with fiber,3G/4G and Satellite connectivity. Kindly if you can advise, if the proposed design is something that can work or the combination of star/mesh topology will create issues when operating on Layer 2 (e.g. Dublicate Multicast Packets).


May I know 1000 radios is equivalent to 1000 BR1s?

1 Like


1000 RF radios is what you are looking to have in our RF and IP Domain. At project maturity we estimate approximately 50 BR1, 15 x 380 and 1 Balance 710 for the multicasting/field units. For the RF radio repeaters, we are looking at a layer 3 global network and approximately another 15 x BR1’s and 1 x Balance 380.

Gents kind reminder, if the proposed setup will work or the said topology will create issues as Dublicate Multicast Packets and/or propagation of ethernet frames

I agree with what Kevin was suggested here.

If full mesh is necessary and Balance 710 is chosen as the root bridge, please take note of the latency and packet loss between Balance 710 and BR1s.

1 Like


I have send you a personal message with my contacts in case you need any help on L2 VPN.