Although RIP has been a mainstay of internetwork routing for quite a long time, there are still some concerns and limitations that need to be addressed. These are as follows:
Inability to support paths with more than 15 hops
(Static) Fixed metrics for route calculation
Network bandwidth consumption for routing table updates
Slow convergence
No dynamic load-balancing support
These issues are discussed in the following sections.
RIP was designed to support networks of relatively small diameter, enforcing a strict hop count of 15 hops maximum, with 16 hops being marked as unreachable.
NOTEThe network diameter is categorized as the longest path, measured in hops, between two end-nodes. |
As data packets are forwarded across a RIP router, their hop counters are incremented by the cost of the link over which they traversed; usually, 1 is the default. If the data packet's hop counter hits 15 and the packet is not at its intended destination, it is dropped and the destination is considered unreachable.
RIP cannot update its cost metrics in a real-time environment to adapt to changes in the network topology. RIP metrics can be changed only manually and are static for the duration. It is because of these fixed, static metrics that RIP is not ideal in supporting real-time applications.
RIP routers broadcast their entire routing table out to every RIP-enabled interface every 30 seconds, rather than sending an update of the affected route. In large internetworks, this can consume a fair amount of network bandwidth that would otherwise be used to carry data traffic.
RIP routing updates are sent every 30 seconds, which is nearly an eternity in "network time." Several things could happen during this 30-second time, such as data traffic being transmitted, or another change occurring in the network topology. Additionally, a single RIP router takes up 180 seconds (three minutes) to invalidate a route. As the network topology grows, this convergence time also grows to a nearly unmanageable state.
This convergence time poorly supports real-time networked applications.
RIP does not have the capability to dynamically load balance across two or more links. If RIP learns of a 56 Kbps path to a destination first, and learns of a T1 Mbps path to the same destination in a later update, RIP will continue to use the 56 Kbps path because that was the first one learned. A RIP router will only use the T1 path if the 56 Kbps path is marked as unreachable (due to a link failure or the like).