Why doesn't BGP... -Reply

Vadim Antonov avg at pluris.com
Wed Nov 20 19:58:11 UTC 1996


Curtis Villamizar <curtis at ans.net> wrote:

>IP routing has an scaling advantage over call setup.  Call setup can
>take advantage of the lowest loaded path at the time of setup,
>providing a better chance to load balance (ala PNNI).

The Pluris load-balanding approach allows gradual shifting of load
between pipes while doing native IP routing.

>The VC table
>has a speed forwarding advantage over the IP radix tree.

The upper levels of the radix tree fit nicely in L1 cache,
so the real advantage of VC table is far smaller.

>The radix
>tree may have a space advantage due to aggregation.

Radix tree is O(log n) from size of the network.  (n is the
number of hosts).  VC table is O(n), at least at some switches.

>That in a
>nutshell is the IP vs ATM flamewar (did I miss anything - ommisions
>private mail unless it was major).

Nah, it is much funnier.  I plan to publish my comments on the issue
at Pluris web site shortly.

>The interest in IP switching
>(CSRs, tag switching, IP switching proposal de jour) comes from a
>desire to combine the best of the two.

What happened to K.I.S.S.?  :)

--vadim





More information about the NANOG mailing list