From: Janet Davis (jlnd_at_cs.washington.edu)
Date: Mon Feb 09 2004 - 10:58:56 PST
A clarification and a questions on hw2 q1:
Clarification: Where it says "when forwarding a packet, each node sends it
to the last neighbor that forwarded a packet from that source address",
read, "last neighbor that forwarded a _previously unseen_ packet from
that source address." That is, you don't update the forwarding table when
you get a duplicate packet.
On Sun, 8 Feb 2004, Tsz (Oscar) Ng wrote:
> For 1a, do we assume that the network topology doesn't change while the
> packet is transmitting (i.e. a node is added or removed)?
Yes, please assume that the topology doesn't change for part (a).
However, you may assume that if two paths have the same propagation delay
to a particular node, two copies of the packet delivered along both paths
will arrive in random order. For instance, consider this topology with
all propagation delays equal:
B
/ \
A D
\ /
C
If A floods a packet to the broadcast address, it's random whether the
first copy to arrive at D will come from B or C.
Cheers,
Janet
-- Janet Davis jlnd_at_cs.washington.edu http://www.cs.washington.edu/homes/jlnd/ _______________________________________________ Cse461 mailing list Cse461_at_cs.washington.edu http://mailman.cs.washington.edu/mailman/listinfo/cse461
This archive was generated by hypermail 2.1.6 : Mon Feb 09 2004 - 10:58:58 PST