RE: FWs in hardware (was: FW-1: Questions blah blah blah)

["KirkAdams" <Kirk-Adams () email msn com>]

This is likely due to "speed" requirements. I spoke with Xylan and suspect
this implementation exists in some of their gig+ switches with HUGE
backplane throughputs.

See my post on "firewalls limited to protecting Ethernet"

Kirk Adams

> I'm aware of at least three implementations where FW-1 runs in hardware:
> from Berkeley Networks, Ipsilon, and Xylan. I've heard the Ipsilon (now
> Nokia) and Xylan implementations run FW-1 on a CPU-based board, but I
> haven't verified this; the Berkeley box downloads FW-1 rule sets onto an
> ASIC.

<snip>
> David Newman
> Data Communications magazine
>
> > He may have meant this, from http://www.ipsilon.com/products/:
>
> [snip]
>
> > > FW-1 in hardware?
> > >
> > > Now I'm really starting to have doubts about this story.