Date: Thu, 16 Jan 1997 01:43:00 +0000 From: Brian Somers <brian@awfulhak.demon.co.uk> To: Archie Cobbs <archie@whistle.com> Cc: julian@whistle.com, danny@panda.hilink.com.au, ejs@bfd.com, freebsd-hackers@freebsd.org, archie@alpo.whistle.com Subject: Re: FreeBSD as an ISDN Router Message-ID: <199701160143.BAA15211@awfulhak.demon.co.uk> In-Reply-To: Your message of "Wed, 15 Jan 1997 16:59:18 PST." <199701160059.QAA23303@bubba.whistle.com>
next in thread | previous in thread | raw e-mail | index | archive | help
[.....]
> > Do you mean that mpd isn't configurable as a single-line ppp daemon ?
> > I was thinking along the lines of it being a user-land ppp that can sit
> > on a variable number of serial lines.
>
> No, what I meant was that mpd doesn't have Charles Mott's alias stuff in it.
So if the alias stuff was merged, it would be a good replacement for
/usr/src/usr.sbin/ppp ? I don't mind which way around the merge is done...
Armed with this sort of beast, it'll be interesting trying to get my ISP to
co-operate and give me a twice-the bandwidth-for-twice-the-price deal ;)
> > BTW, how does mpd do its load balancing ? Does it just send packets to
> > the interface with the smallest buffer queue ?
>
> No, it uses a fairly simple strategy whose goal is to make all of the
> packet fragments arrive on the other end of the link(s) at the same time.
> It computes this based on user inputted values for latency and bandwidth.
> This optimizes for both total bandwidth and total latency.
This would be a convenient way to test that the alias stuff deals with out-of-order fragments properly too (this hasn't been tested properly 'cos we can't easily create the environment) :)
So, are there any objections to this (merging) ?
--
Brian <brian@awfulhak.demon.co.uk>, <brian@freebsd.org>
<http://www.awfulhak.demon.co.uk/>;
Don't _EVER_ lose your sense of humour....
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199701160143.BAA15211>