David Newman wrote:
> >
> > > Cars slow down when approaching and toll booth speed up going
> > away from it,
> > > and that affects their "throughput." Ditto packets traversing firewalls.
> >
> > Not if the acceleration lanes are wide enough: 20 lanes of
> > traffic moving at
> > 10 MPH has the same throughput as 5 lanes of traffic moving at 40 MPH.
> > Similarly, a "full speed" firewall may need to have several NICs
> > on each side.
> > Parallelism solves many throughput problems, but rarely benefits latency
> > (except for reduced queue length).
>
> Eh? Here the analogy breaks. Regardless of the number of lanes, ALL the
> cars/packets were going 65 mph before they hit the toll booth/firewall. You
> need a hell of a lot of parallelism to make up for that.
You need a precisely measurable amount of parallelism to handle that. If the
cars go from 65 MPH to 6.5 MPH (on average through the toll gate) then you need
to go from 2 lanes to 20 lanes. Is that "a hell of a lot"? Sure, it's more
than most toll plazas that I've ever seen, but most traffic authorities are not
so concerned with throughput that they will engineer a full-bandwidth toll
plaza under peak load.
Similarly, most firewall vendors/customers are not so concerned with throughput
that they will pay the (substantial) cost of a machine with enough
computes/parallelsim to do sophisticated inspection at full network bandwidth.
So it's rare and expensive, but not impossible
Crispin
-----
Crispin Cowan, CTO, WireX Communications, Inc. http://wirex.com
Free Hardened Linux Distribution: http://immunix.org
JOBS! http://immunix.org/jobs.html
Received on Mar 21 2000
Intro
