522 views
0 votes
0 votes

I have a little confusion about some concepts in Distance Vector Routing. Please help. 

Is it correct to say the following, with regard to Distance Vector routing:


1. Any finite change to the cost of a link will be updated INSTANTANEOUSLY onto the routing table of the nodes connected by the link. For example in the below diagram if the cost of link N2-N3 changes to 2,

then the routing tables of both N2 and N3 will be updated to:

N2: (1, 0, 6, 7, 3) -> N2: (1, 0, 2, 7, 3)

N3: (7, 6, 0, 2, 6) -> N3: (7, 2, 0, 2, 6)

2. In the next round of updates (sharing of routing tables), the vectors sent by N2 and N3 WILL CONTAIN the new values due to the change in cost of the link N2-N3. 

The answer to question https://gateoverflow.in/2160 validates the above two points.  However, the answer to this one https://gateoverflow.in/43317 invalidates them. How? Via this statement: 

Now for N3 to get updated in the subsequent round it will get tables from N2 and N4 only. But first we need to find the N4 calculated in previous update. So in previous question N4 received updates from N3 and N5 which are N3: (7, 6, 0, 2, 6),N5: (4, 3, 6, 4, 0).

The reduction of cost of the link N2-N3 is not taken into consideration in the routing table of N3.

I would be grateful if someone clarified this doubt. Below is my reasoning, read only if you want to. :)

My Logic:

In distance vector routing, the sharing of vectors by routers are done parallelly, i.e. at the same instance. For this very reason, while calculating, we consider the old values of the individual routing tables and not the new ones constructed as a result of the sharing of tables.

But if a change in the cost of the link is updated onto the adjucent tables (N2,N3) instantaneously, the vectors sent by these routers, in the next round of exchange, will contain the new values due to the change in cost of the link.  

Please log in or register to answer this question.

Related questions

0 votes
0 votes
0 answers
4