From owner-freebsd-net Wed Jan 24 19:53:31 2001 Delivered-To: freebsd-net@freebsd.org Received: from prism.flugsvamp.com (cb58709-a.mdsn1.wi.home.com [24.17.241.9]) by hub.freebsd.org (Postfix) with ESMTP id 8350237B401 for ; Wed, 24 Jan 2001 19:53:14 -0800 (PST) Received: (from jlemon@localhost) by prism.flugsvamp.com (8.11.0/8.11.0) id f0P3qpV49010; Wed, 24 Jan 2001 21:52:51 -0600 (CST) (envelope-from jlemon) Date: Wed, 24 Jan 2001 21:52:51 -0600 (CST) From: Jonathan Lemon Message-Id: <200101250352.f0P3qpV49010@prism.flugsvamp.com> To: genuine@ma4.justnet.ne.jp, net@freebsd.org Subject: Re: estimation of RTT,RTO from tcpcb parameter X-Newsgroups: local.mail.freebsd-net In-Reply-To: References: Organization: Cc: Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In article you write: >Thank you very much for your kindly advise. > >>The system timer ticks at a rate of 100 Hz on all IA32 processors. (I >>believe on Alphas it ticks at 1024 Hz but I may be imagining it.) >>Thus, the period is either 10 ms or 0.98 ms. >> >>In historic versions of FreeBSD, most TCP timers ticked at 2 Hz while >>the others ticked at 5 Hz. This was changed in advance of FreeBSD >>4.0. > >I see. >Anyway, it seems to be difficult that estimating RTT, RTO within a LAN >envirounment (such as Gigabit connection). >Of course, this time scale may be valid for general internet envirounment. >I'll study more. The math for RTT calculations is explained somewhere in Van Jacobsen's congestion avoidance & control paper, I believe. The equations have since been modified slightly since then, but give the basic idea. -- Jonathan To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message