A proposal from Cisco engineers is spawning a new IETF working group to explore scalability issues associated with multihoming, colleague Carolyn Duffy Marsan reports. The Cisco proposal, called Locator/Identifer Separation Protocol (LISP), describes a method for reducing the number of entries in the BGP routing tables of core routers when enterprises split traffic among multiple carriers.
LISP proposes a new tunneling mechanism to be used by the Internet's edge and core routers, Marsan reports. The protocol logically separates a block of IP addresses that a company advertises into two functions: one for identifying the systems using the IP addresses; and the other for locating where these systems connect to the Internet.
This separation allows LISP-enabled edge routers to aggregate the location information, so less of it needs to be stored in the core routers.
But skeptics say LISP actually adds overhead and complexity to core routers by requiring processing of mapping and tunneling services. That's one of the reasons the IETF plans to keep LISP an experimental protocol for now.
It also sounds a bit like MPLS, no? LISP does have some traffic engineering (TE) capabilities, proponents say; but as I understand it, LISP is specific to multihoming and lives between the customer edge and the provider edge, while MPLS TE Forwarding Information Bases (FIBs) are exchanged between provider and core systems.
So the Internet facing edge is telling FIBs, and the customer facing edge is speaking with LISPs. No wonder it's hard to guarantee anything on the Internet...
More from Cisco Subnet:Cisco picks partners for next-gen data center assaultDow Jones Industrial Average: Cisco may have the right stuffAre Wireless Certifications Right For You?Cisco cert lab prep: Sim vs Em vs RealCisco taking share from Nortel?IT in 2109: Living Quantum Computers, Mind Reading PCs, and DNA nanofiber CPUsWendell Odom's OSPF training series