This preview shows page 1. Sign up to view the full content.
Unformatted text preview: the total bandwidth consumed is
(4km2 + 2mn)/ADVERT INTERVAL bytes/second.
It is easy to see that there is no connected network in which the bandwidth consumed
by the simple link-state protocol is lower than the simple distance-vector protocol; the
important point is that the former is quadratic in the number of links, while the latter
depends on the product of the number of nodes and number of links.
Convergence time. The convergence time of our distance vector and path vector protocols can be as large as the length of the longest minimum-cost path in the network multiplied by the advertisement interval. The convergence time of our link-state protocol is
roughly one advertisement interval.
Robustness to misconﬁguration. In a vector protocol, each node advertises costs and/or
paths to all destinations. As such, an error or misconﬁguration can cause a node to wrongly
advertise a good route to a destination that the node does not actually have a good route
for. In the worst case, it can cause all the trafﬁc being sent to that destination to be hijacked
and possibly “black holed” (i.e., not reach the intended destination). This kind of problem
has been observed on the Internet from time to time. In contrast, the link-state protocol
only advertises each node’s immediate links. Of course, each node also re-broadcasts the
advertisements, but it is harder for any given erroneous node to wreak the same kind of
havoc that a small error or misconﬁguration in a vector protocol can.
In practice, link-state protocols are used in smaller networks typically within a single
company (enterprise) network. The routing between different autonomously operating
networks in the Internet uses a path vector protocol. Variants of distance vector protocols
that guarantee loop-freedom are used in some small networks, including some wireless
“mesh” networks built out of short-range (WiFi) radios. Acknowledgments Thanks to Sari Canelake for several useful comments. CHAPTER 19. NETWORK ROUTING - II 12 ROUTING AROUND FAILURES Problems and Questions
1. Why does the link-state advertisement include a sequence number?
2. What is the purpose of the hop limit ﬁeld in packet headers? Is that ﬁeld used in
routing or in forwarding?
3. Describe clearly why the convergence time of our distance vector protocol can be as
large as the length of the longest minimum-cost path in the network.
4. Suppose a link connecting two nodes in a network drops packets independently with
probability 10%. If we want to detect a link failure with a probability of falsely reporting a failure of ≤ 0.1%, and the HELLO messages are sent once every 10 seconds,
then how much time does it take to determine that a link has failed?
5. *PSet* You’ve set up a 6-node connected network topology in your home, with nodes
named A, B, . . . , F . Inspecting A’s routing table, you ﬁnd that some entries have been
mysteriously erased (shown with “?” below), but you ﬁnd the following entr...
View Full Document
- Fall '13