* Could Layer 3 MPLS VPNs work for you?
endif; ?>Recently, a reader wrote to tell me about the big relief he felt when he bought a Layer 3 MPLS VPN service. With such a site-to-site service, one option is to peer the WAN routers at customer sites directly to the edge routers of the service provider, and that’s it. The customer has no responsibility for routing directly among his own WAN routers..
Recently, a reader wrote to tell me about the big relief he felt when he bought a Layer 3 MPLS VPN service. With such a site-to-site service, one option is to peer the WAN routers at customer sites directly to the edge routers of the service provider, and that’s it. The customer has no responsibility for routing directly among his own WAN routers.
The larger the number of sites, the more work this relieves because the number of peering relationships soars as each additional site is added.
The reader was actually switching from a frame relay network to MPLS and was going from a partially meshed network to a fully meshed network because of both price and complexity. It was far cheaper to buy the MPLS service than to buy more frame relay virtual circuits to mesh the network.
Some service providers offer tiers of service quality so customers can pay more for certain traffic to get better treatment. An offshoot of this is the ability to get service level agreements out of providers that exact penalties or credits if quality promises are not met.
Despite these advantages, Layer 3 MPLS VPNs might not be for everybody. In particular, those with policies that state they must be in charge of their WAN routing for security reasons won’t be interested.
Also for security reasons, potential users might prefer a Layer 2 service that would inherently maintain a certain level of protection. Otherwise, they would have to overlay security via separate authentication or perhaps a traditional IPSec VPN, which would be secure but would also add work and cost.
If the downside of the services are not an obstacle, Layer 3 MPLS VPNs are a flexible and potentially cost-effective option.




