[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [RRG] FIB cache?
András,
> On the Saturday seminar there was a discussion that only a few prefixes are responsible for majority of the traffic, while a lot of prefixes generate no or very minimal traffic. Someone had a comment that this speaks for having a FIB cache.
Most of the papers that studied the Internet traffic characteristics
have found that a small number of prefixes are responsible for a large
fraction of the traffic. This is nice if we expect that a FIB cache will
be used to forward the packets towards those *heavy* prefixes. However,
if the router needs to forward packets towards any prefix, then the
situation is less favourable as the measurements also show that a router
sends a small amount of packets (e.g. one packet from time to time) to a
large number of prefixes.
In 1999, we did an analysis based on Netflow data showing the number of
prefixes used as destination by two ISP networks : a small dialup
provider and a research network. At that time, the Internet had 70K BGP
prefixes. See :
S. Uhlig and O. Bonaventure. On the cost of using MPLS for interdomain
traffic. In Quality of Future Internet Services (QoFIS 2000), September 2000
http://totem.info.ucl.ac.be/publications/papers-elec-versions/Uhlig-Bonaventure-QOFIS2000.pdf
Olivier
--
to unsubscribe send a message to rrg-request@psg.com with the
word 'unsubscribe' in a single line as the message text body.
archive: <http://psg.com/lists/rrg/> & ftp://psg.com/pub/lists/rrg