Subject: Re: ifaliases
To: Jason Thorpe <thorpej@NAS.NASA.GOV>
From: Perry E. Metzger <firstname.lastname@example.org>
Date: 06/20/1996 14:55:52
Jason Thorpe writes:
> Improving the performance of the traversal probably wouldn't be that
> hard to do; the address could be hashed,
This is what I favor...
> Another way to improve the performance of the address list is to use
> the notion of temporal locality to dynamically re-sort the list.
This will likely work well in small situations, as LAN traffic at low
bandwidths tends to be pretty good for locality of reference, but it
may not work as well on machines like multicast routers. The higher
the traffic, the lower the likely locality of reference, which is
exactly when you need performance the most. I'd say to try the
simplest method, which is hashing. It will likely perform the best
over a wide range of network conditions.