From owner-freebsd-hackers Wed Feb 15 10:00:55 1995 Return-Path: hackers-owner Received: (from root@localhost) by freefall.cdrom.com (8.6.9/8.6.6) id KAA28569 for hackers-outgoing; Wed, 15 Feb 1995 10:00:55 -0800 Received: from brasil.moneng.mei.com (brasil.moneng.mei.com [151.186.20.4]) by freefall.cdrom.com (8.6.9/8.6.6) with SMTP id KAA28563; Wed, 15 Feb 1995 10:00:53 -0800 Received: by brasil.moneng.mei.com (4.1/SMI-4.1) id AA18784; Wed, 15 Feb 95 11:58:11 CST From: Joe Greco Message-Id: <9502151758.AA18784@brasil.moneng.mei.com> Subject: Re: Network gurus: How hard to split bandwidth across modems? To: tom@haven.uniserve.com (Tom Samplonius) Date: Wed, 15 Feb 1995 11:58:10 -0600 (CST) Cc: jcargill@cs.wisc.edu, jkh@FreeBSD.org, freebsd-hackers@FreeBSD.org In-Reply-To: from "Tom Samplonius" at Feb 14, 95 06:39:04 pm X-Mailer: ELM [version 2.4beta PL9] Content-Type: text Content-Length: 1562 Sender: hackers-owner@FreeBSD.org Precedence: bulk > > One thing I'm wondering, though; would an implementation of > > Multilink-PPP talk happily to the load-sharing stuff that a NetBlazer > > implements? Or is that a proprietary thing? Does anyone know the > > details on what their bandwidth splitting does? (In case you haven't > > guessed yet, the other end of my connection is a blazer... Thus my > > high degree of personal interest in what it implements... ;-) > > After having much grief with two Netblazer (a PN-2 and a ST), running a > load-balanced SLIP link over two 28.8k modems. Netblazer's use a very > simple algorithm, if a output buffer on the first interface exceeds a > certain level, try the next interface (on a packet by packet basis). The > maximum buffer levels are configurable. Well, it is immediately obvious that they've used a simplistic algorithm: I suspect that it would be No Big Deal to talk to a blazer in whatever manner of multi-line implementation we choose... In response to Jon's previous message, I concur that a multi-link PPP implementation would be preferable. However, given that I am not much of a kernel hack, and that there is code that already works (and just needs to be ported/hacked) for SLIP, I think I would prefer to tackle the job that looks mildly intimidating, rather than the job that looks hopelessly impossible. grin :-) ... Joe ------------------------------------------------------------------------------- Joe Greco - Systems Administrator jgreco@ns.sol.net Solaria Public Access UNIX - Milwaukee, WI 414/342-4847