From owner-freebsd-isp Sat May 17 07:29:42 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id HAA25346 for isp-outgoing; Sat, 17 May 1997 07:29:42 -0700 (PDT) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA25341 for ; Sat, 17 May 1997 07:29:40 -0700 (PDT) Received: from ntws (db@ntws.etinc.com [204.141.95.142]) by etinc.com (8.8.3/8.6.9) with SMTP id KAA15065; Sat, 17 May 1997 10:38:53 -0400 (EDT) Message-Id: <3.0.32.19970517102841.00b2d9e0@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Sat, 17 May 1997 10:28:45 -0400 To: Tony Li , Bradley Dunn From: dennis Subject: Re: interface card to connect 64k..256k to connect to internet Cc: isp@FreeBSD.ORG Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-isp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk At 11:26 PM 5/16/97 -0700, Tony Li wrote: >bradley@dunn.org (Bradley Dunn) writes: > >> ASUS P6NP5 w/ 150-MHz Pentium Pro: $435 (5 PCI slots) >> Intel Etherexpress Pro 10/100 PCI: $79 >> >> Now the question of at what point this router will saturate the PCI bus is >> more interesting... > >That particular combination can't saturate it. In fact, even if you fill >the PCI bus with 4xT1 cards, you have plenty of bandwidth: > >4XT1 = 4x(1.5Mbpsx2) = 12Mbps >4 cards@12Mbps = 48Mbps >1x 10/100 = 100Mbps > >Total=148Mbps. PCI can (and does) deliver up to 500Mbps. Of course we were talking about 4 QUAD T1 cards, but you can do the math.... ISA can handle 6 T1 lines in practice...and worst case PCI is 8 times faster, usually MUCH better than that. The difference is that with ISA, it IS the limiting factor, with PCI, its the OS processing as you pps get very high that is the limiting factor. Dennis