Talk:Distance-vector routing protocol

Unclear and confusing

Paragraphs in this article might be confusing or difficult to understand for some readers. For example,

"...This slowly propagates through the network until it reaches infinity (in which case the algorithm corrects itself, due to the relaxation property of Bellman–Ford)."

The idea of something being able to reach infinity would be confusing to many readers without additional background. This paragraph should be reworded so that the meaning of the term, "infinity" is made clear. The words, "... algorithm corrects itself due to relaxation property of Bellman-Ford" is also difficult to understand. I would encourage review into this paragraph. Thanks for your assistance.

--Hrbm14 (talk) 12:56, 14 September 2013 (UTC)

Yellow vs. Green

your definition of "yellow" in the example is more close to my definition of "green" than "yellow" —Preceding unsigned comment added by 71.150.249.231 (talk) 22:12, 23 March 2009 (UTC)

Example

replaced IGRP example by RIP example... —Preceding unsigned comment added by The Anome (talk • contribs) 10:07, 4 January 2004

Need to cover

This page should be expanded to cover what Yakov named "path-vector" algorithms, which are part of a large class of "destination vector" algorithms, along with "distance-vector" (in that they both pass their neighbours vectors of destination info, i.e. routing tables). Noel (talk) 15:26, 2 October 2005 (UTC)

More explanation

This article seems to be good, (if a tad short) but the graph at the bottom needs some explanation. For example, what is T? Unless I am mistaken, it doesn't say in the article. Someone who knows about this stuff needs to add a bit more to this. Spacerat3004 23:28, 14 February 2007 (UTC)

I fleshed out the example, still missing a better explanation for poison reverse, and why it doesn't work for all cases. (I think it only works for loops of 3 routers, or of a "diameter" of 2. right?) --Scraimer 00:43, 23 February 2007 (UTC)

less message overhead

"Compared to link-state protocols, which requires a router to inform all the nodes in a network of topology changes, distance-vector routing protocols have less computational complexity and message overhead." - is this really true? Link state is generally considered to scale better as its link state packets are far smaller than distance vectors routing info packets —Preceding unsigned comment added by 128.232.108.57 (talk) 20:58, 27 May 2008 (UTC)

No, pure link state actually scales pretty badly. OSPF works around that by splitting the network into areas, and performing distance vector routing between the areas. Jec (talk) 19:49, 29 May 2011 (UTC)

Nonetheless, the message overhead of a normal DV routing protocol tends to be bigger than the one from Link-State Protocols. This happens when the DV protocol uses (1) periodic updates; and (2) full routing table updates. Both disavantages are present in RIP, which is the first and main example of DV routing protocol. Interenstingly, neither of these disavantages are present in EIGRP (which is called an Advanced DV routing protocol). It's also important to differ "message overhead" from "scaling"; and not forget that the concept of using areas is from the L-S paradigm, not from OSPF. — Preceding unsigned comment added by Zekkerj (talk • contribs) 19:08, 18 August 2011 (UTC)

Reaching infinity

reaches infinity (in which case the algorithm corrects itself [...])

What does this mean? Surely something cannot reach infinity. If you could reach it, it would be finite. --Ysangkok (talk) 03:49, 5 February 2012 (UTC)