Date: Fri, 13 Jul 2001 14:45:24 -0400 From: Leo Bicknell <bicknell@ufp.org> To: Julian Elischer <julian@elischer.org> Cc: Leo Bicknell <bicknell@ufp.org>, freebsd-hackers@freebsd.org Subject: Re: Network performance roadmap. Message-ID: <20010713144524.A24595@ussenterprise.ufp.org> In-Reply-To: <Pine.BSF.4.21.0107131302240.69775-100000@InterJet.elischer.org>; from julian@elischer.org on Fri, Jul 13, 2001 at 01:14:07PM -0700 References: <20010713101107.B9559@ussenterprise.ufp.org> <Pine.BSF.4.21.0107131302240.69775-100000@InterJet.elischer.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, Jul 13, 2001 at 01:14:07PM -0700, Julian Elischer wrote: > I would suggest that 2 * the current window may be too small > because the window might be increasing and > reception of one ack might move the window up by > the entire transmitted window size, resulting in starvation > if the waiting data can be sent quickly. therefore I suggest a few > modifying factors: Actually, I doubt 2 * cwin would be too small, and here's my rational: * The majority of the "large increases" should be during slow start. These values should all be below the minimum buffer size, so there is no thrashing. Note, this requires a good minimum buffer size. * When dealing with large bandwidth*delay number (say, a 1 meg cross country 100Mbps path) we don't really want to buffer 3 meg, rather than 2 meg. In fact, I could question if we really want 2 meg, but it's an ok compromise for now. -- Leo Bicknell - bicknell@ufp.org Systems Engineer - Internetworking Engineer - CCIE 3440 Read TMBG List - tmbg-list-request@tmbg.org, www.tmbg.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010713144524.A24595>