The MPLS-OPS Archive[Date Prev][Date Next][Thread Prev][Thread Next] [Date Index][Thread Index][Author Index][Subject Index] RE: MPLS L2 VPN end L3 VPN
I am not quite sure what do you mean by the scaling problem associated with a Layer 2 VPN. Since the PE routers are no longer peering with the CE routers, the need for separate VRFs is eliminated leading to a more scalable implementation of a VPN. Further, I agree that the Layer 2 implementation of a VPN is limited to a specific transport medium but there do exist provisions for constructing a Layer 2 VPN with two different layer 2 networks at the two ends of the PseudoWire. I am told that these features are vendor specific. I know for a fact that Cisco boxes do not support this (at least, not yet!) Ravi Bhagavathula Wichita State University -----Original Message----- From: Sonali Francis [mailto:Sonali.Francis@lntinfotech.com] Sent: Thursday, March 27, 2003 7:20 AM To: Alexandre Carlos de Oliveira Cc: mpls-ops@mplsrc.com Subject: Re: [MPLS-OPS]: MPLS L2 VPN end L3 VPN Here are some ideas on Layer 2 and Layer 3 VPN: The Layer 2 VPN is mainly based on overlay model where PE maps maps incoming Layer 2 traffic onto the appropriate point-to-point tunnel. Simple point-to-point tunnels are established on an provider's network to handle various forms of Layer 2 traffic (Ethernet, frame relay, ATM, TDM, and PPP/HDLC). From the peering aspect, CE router is peer to CE router; PE is not a peer to the CE and does not maintain separate routing tables. No L3 processing of the customer packet takes place in PE. The Advantages are 1. Advantageous where direct interoperability with existing Layer 2 VPN deployments is important. 2. Layer 2's MPLS "overlay" model is more attractive for carriers with existing Layer 2 VPN Disadvantages are 1. Scaling problem: Lack the scope of routed networks, limiting a Layer 2 implementation to the confines of the transport medium The Layer 3 VPN is based on Peer-to-peer Model. In case of Layer 3 MPLS VPN, it is implemented by using two-level label stack. The ingress PE router pushes both a Next-Hop BGP header (for the private network) and a Next-Hop Interior Gateway Protocol (IGP) header (for the shared infrastructure) onto the packet. After reaching the egress PE router via one or more MPLS Label Switched Paths (LSPs), the PE pops the MPLS headers and delivers a normal IP packet to the customer. In this case CE routers peers with PE routers. The advantages are 1.Ubiquity of IP networks over multiple transport networks. 2. Multiple customers having IP running over different data link layer can form a VPN. 3.Dynamic VPNs by virtue of strong automatic route discovery. 4. Economical for ISPs that already use BGP extensively and have already deployed high-end IP/MPLS routing equipment at the edge. But Disadvantage is this is more complex and expensive http://lambdanet.net/userfiles/france/download/white_paper/Keep_it_Simpl e_with_BGP_MPLS_VPN_LambdaNet.pdf is quite informative. But it is to be noted that Layer 2 or layer 3 VPN can exist in non-MPLS domain too. Please correct me if I am wrong.... Regards, Sonali Alexandre Carlos de Oliveira To: <mpls-ops@mplsrc.com> <alexandre@redes. cc: unb.br> Subject: [MPLS-OPS]: MPLS L2 VPN end L3 VPN 26/03/2003 11:16 PM Can some one tell me whats the difference between L2 VPN and L3 VPN? Alex ------- The MPLS-OPS Mailing List Subscribe/Unsubscribe: http://www.mplsrc.com/mplsops.shtml Archive: http://www.mplsrc.com/mpls-ops_archive.shtml ------- The MPLS-OPS Mailing List Subscribe/Unsubscribe: http://www.mplsrc.com/mplsops.shtml Archive: http://www.mplsrc.com/mpls-ops_archive.shtml ------- The MPLS-OPS Mailing List Subscribe/Unsubscribe: http://www.mplsrc.com/mplsops.shtml Archive: http://www.mplsrc.com/mpls-ops_archive.shtml
|
|