From owner-freebsd-stable Thu Jul 11 16:25:42 2002 Delivered-To: freebsd-stable@freebsd.org Received: from mx1.FreeBSD.org (mx1.FreeBSD.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 5E5F537B400 for ; Thu, 11 Jul 2002 16:25:40 -0700 (PDT) Received: from HAL9000.wox.org (12-233-156-170.client.attbi.com [12.233.156.170]) by mx1.FreeBSD.org (Postfix) with ESMTP id AFD6343E3B for ; Thu, 11 Jul 2002 16:25:39 -0700 (PDT) (envelope-from dschultz@uclink.Berkeley.EDU) Received: from HAL9000.wox.org (localhost [127.0.0.1]) by HAL9000.wox.org (8.12.3/8.12.3) with ESMTP id g6BNPkGR001551; Thu, 11 Jul 2002 16:25:46 -0700 (PDT) (envelope-from dschultz@uclink.Berkeley.EDU) Received: (from das@localhost) by HAL9000.wox.org (8.12.3/8.12.3/Submit) id g6BNPema001549; Thu, 11 Jul 2002 16:25:40 -0700 (PDT) (envelope-from dschultz@uclink.Berkeley.EDU) Date: Thu, 11 Jul 2002 16:25:40 -0700 From: David Schultz To: Thierry Herbelot Cc: Matthew Dillon , freebsd-stable@FreeBSD.ORG Subject: Re: tuning(7) request was: Re: Performance boost with kernel options in FBSD 4.6 Message-ID: <20020711232540.GA1437@HAL9000.wox.org> Mail-Followup-To: Thierry Herbelot , Matthew Dillon , freebsd-stable@FreeBSD.ORG References: <20020710104730.L10343-100000@klima.physik.uni-mainz.de> <04a601c228dc$c6dbb980$681663cf@icarz.com> <200207111930.g6BJUX5m096974@apollo.backplane.com> <3D2DF09C.B553DE07@herbelot.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <3D2DF09C.B553DE07@herbelot.com> Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG Thus spake Thierry Herbelot : > Matthew Dillon wrote: > > > > An increased switching rate (increasing HZ) may be useful in the above > > situation. Still, I would not recommend increasing Hz above 500 (2ms). > > 10000 (100uS) is just plain insane. > > > > from actual experience, any p-III with a clock rate above 500MHz (that > is, any recent CPU) can sustain Hz=5000, which I used to run > trafific-shaped packet blasters (admittedly a narrow focus ...) with > very good results (better than special-purpose test boxes). > > As is said in the dummynet man page, FreeBSD can be a very good traffic > shaper, if the userland scheduling rate is high enough (will it be the > same with threads in -current, with KSE ?). Sure modern processors can handle 5000 Hz at the expense of a few CPU cycles, but do you actually find that dummynet isn't accurate enough at 1000 Hz? To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message