Message info
 
To:Christopher Morrow From:Brian Dickson Subject:Re: [sidr] Keys and algorithms for Updates - feasibility analysis? (was Re: RPKI and private keys) Date:Mon, 7 May 2012 17:00:55 -0400
 

Even at CPU parity, or even with modest N x CPU performance for single digit N, same ballpark.
The expected performance once you load up the number of BGPSEC sessions, add on churn, etc., is a lot worse.
And, my little laptop was basically idle, not doing all the other work that a router CPU needs to do.

Brian

On Mon, May 7, 2012 at 4:52 PM, Christopher Morrow <morrowc.lists@gmail.com> wrote:
On Mon, May 7, 2012 at 2:58 PM, Brian Dickson
<brian.peter.dickson@gmail.com> wrote:
> And given that current generation hardware has CPUs at least an order of
> magnitude slower, or possibly two orders of magnitude, suggests that
> software-based bgpsec can never work.

for clarity, I think a 'normal':
o 7200 these days ships with a 1ghz cpu
o Cisco CRS ~2.13ghz Intel
o Juniper M/T/MX ~2ghz Intel
o Cisco 12000/XR (prp2 single-core, 3 dual-core) 1.3GHz PPC

None of these is an order of magnitude smaller than a desktop-cpu.
Even a 7500 with a 600mhzRISC is not an order of magnitude. (by the
mhz/ghz numbers at least, speedups in math done in assembly aside)