Routing Tables and Path Selection

---

• most end system workstation and server computers are:
  • configured with a single network adapter
  • connected to only one network
• could be capable of routing but,
  • not equipped with necessary interfaces and knowledge of the location of other networks

A router is a multihomed intermediate system with links and network topology information to facilitate forwarding packets between subnets or around an internetwork.

• information about the location of other IP networks and hosts is stored in a routing table
  • each entry represents an available path to a destination network or host
  • parameters of a routing entry:
    • protocol
      • source of the route
      • paths can be configured:
        • statically
        • learned by exchanging information with other routers via a dynamic routing protocol
    • destination
      • routes can be defined to specific hosts but are more generally directed to network IDs
      • if more than one match
        • most specific (longest) destination prefix will be selected as the forwarding path
    • interface
      • local interface to use to forward a packet along the chosen route
      • represented as:
        • IP address of the interface
        • or layer 2 interface ID
    • gateway/next hop
      • IP address of the next router along the path to the destination

Info

• IPv4 host has /32 prefix
• IPv6 host has /128 prefix
• directly connected hosts are defined by host routes
• anycast addressing also requires that the anycast address be entered in routing tables as a host route