From owner-freebsd-hackers Sat Jun 24 16:30:07 1995 Return-Path: hackers-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id QAA29582 for hackers-outgoing; Sat, 24 Jun 1995 16:30:07 -0700 Received: from cs.weber.edu (cs.weber.edu [137.190.16.16]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id QAA29576 for ; Sat, 24 Jun 1995 16:30:05 -0700 Received: by cs.weber.edu (4.1/SMI-4.1.1) id AA17185; Sat, 24 Jun 95 16:50:45 MDT From: terry@cs.weber.edu (Terry Lambert) Message-Id: <9506242250.AA17185@cs.weber.edu> Subject: Re: FreeBSD as a router To: nc@ai.net (Network Coordinator) Date: Sat, 24 Jun 95 16:50:45 MDT Cc: davidg@Root.COM, hackers@freebsd.org In-Reply-To: from "Network Coordinator" at Jun 24, 95 04:39:12 pm X-Mailer: ELM [version 2.4dev PL52] Sender: hackers-owner@freebsd.org Precedence: bulk > Second question, and related. Even if an software based router has to > wait for an entire packet frame to come in before routing it, that > strikes me only as a latency problem, and not a thruput problem, Depends on whether another frame is allowed to present until after the incoming frame has been processed. The amount of data you can recieve in the latency period on all your interfaces (up to your routing capacity -- the real limiting factor) dictates the amount of memory you need for buffering. And the more of that you have, the more DRAM refresh cycles you'll need, and the lower your routing capacity. 8-(. Terry Lambert terry@cs.weber.edu --- Any opinions in this posting are my own and not those of my present or previous employers.