<html><head><style type='text/css'>body { font-family: 'Times New Roman'; font-size: 12pt; color: #000000}</style></head><body>Very interesting study I had not seen, and a bummer. That really puts a cramp in my advocation of our CARP+pf load balancers/firewalls/gateways. Than again, what's a PIX box capable of?<br><br>I also had to switch to OpenBSD as there was a fatal crash with the bridge device in FreeBSD when used with my paticular OpenVPN/CARP/pf combination.<br><br>AFAIK pf/forwarding only takes place on one core and wouldn't take advantage of the other 3 cores, correct?<br><br>-Patrick<br><br>----- Original Message -----<br>From: "Adrian Chadd" <adrian@creative.net.au><br>To: "Chris Grundemann" <cgrundemann@gmail.com><br>Cc: "William Herrin" <herrin-nanog@dirtside.com>, nanog@nanog.org<br>Sent: Tuesday, March 25, 2008 6:02:03 PM (GMT-0800) America/Los_Angeles<br>Subject: Re: 10GE router resource<br><br><br>On Tue, Mar 25, 2008, Chris Grundemann wrote:<br><br>> To Ann's question on resources; I have only used Linux routers with 1G<br>> ports but have surpassed 10G total throughput (up+ down) using various<br>> dual proc set ups, most often Intel Xeon in Dell servers. A gentlemen<br>> by the name of Martin Pels wrote a good paper on the subject early<br>> last year that can be found here:<br>> http://docs.rodecker.nl/10-GE_Routing_on_Linux.pdf. He hit a wall at<br>> 700K pps and was using two dual core Intel Xeon 64bit 2.33GHz CPUs and<br>> 2GB of RAM in a Dell PowerEdge 1950.<br><br>Mike Tancsa did some benchmarking in late 2006:<br><br>http://www.tancsa.com/blast.html<br><br>I think things are slightly faster now but not because of a massive<br>change in software architecture.<br><br><br><br><br>Adrian<br><br></body></html>