From owner-freebsd-hackers Thu Jan 31 14:41:13 2002 Delivered-To: freebsd-hackers@freebsd.org Received: from mail.webjockey.net (mail.webjockey.net [208.141.46.3]) by hub.freebsd.org (Postfix) with ESMTP id F083B37B402 for ; Thu, 31 Jan 2002 14:41:09 -0800 (PST) Received: from outloud.org (IDENT:nobody@home.webjockey.net [208.141.46.11]) (authenticated) by mail.webjockey.net (8.11.6/8.11.6) with ESMTP id g0VMZtx46204; Thu, 31 Jan 2002 17:35:55 -0500 (EST) (envelope-from gary@outloud.org) Received: from 208.141.46.249 (proxying for 63.68.129.181) (SquirrelMail authenticated user ancient) by test.outloud.org with HTTP; Thu, 31 Jan 2002 17:36:10 -0500 (EST) Message-ID: <3197.208.141.46.249.1012516570.squirrel@test.outloud.org> Date: Thu, 31 Jan 2002 17:36:10 -0500 (EST) Subject: Re: Clock Granularity (kernel option HZ) From: "Storms of Perfection" To: X-XheaderVersion: 1.1 X-UserAgent: Mozilla/4.0 (compatible; MSIE 6.0; Windows 98; Q312461) In-Reply-To: <3C59C198.78EDFEA9@herbelot.com> References: <3C59C198.78EDFEA9@herbelot.com> X-Priority: 3 Importance: Normal X-MSMail-Priority: Normal Cc: , X-Mailer: SquirrelMail (version 1.2.5 [cvs]) MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG I'm going to benchmark different network senarious with different options to see what I can get, and what works best. If someone wants to help me out, I could maybe write up a article about it? > I've used a large collection of PCs running somewhat real-time network > analysis with a HZ set at 5000Hz with absolutely no ill effects (this > was with P-III-450's) > > using HZ=10000 was outside of the possibilities of the machines. > > one big gain is with timing, which will be better (I myself used NTP to > have a coherent timing on the collection of PC's, with an > inter-correlation better than 1 ms) > > TfH > > Eugene Panchenko wrote: >> >> Hello! >> >> I've seen various postings on the Net where people reported >> network-related and overall performance improvements caused >> by settig HZ kernel option to 1000 (for example), that is, >> reducing a tick size to 1ms for their FreeBSD and Linux >> systems. However, several problems seem to arise, such as >> some device drivers do not include HZ in calculating their >> timeout value, but simply assume HZ to be 100, and also some >> utility programs such as top or ps take timing information >> from the kernel in ticks, also assuming 10ms ticks, however, >> most of these saying were related to Linux. How safe it is >> to bump up HZ to, say, 1000 in FreeBSD (I use 4.5-STABLE)? >> What pitfals will I encounter (drivers, top/ps)? Is there >> are going to be [promised] performance increase? Do I >> really need it? Thank you. > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-hackers" in the body of the message Gary Stanley Network Security Engineer PRECISIONet/Webjockey, Inc. (877) 595-8570 Tickle us, do we not laugh? Prick us, do we not bleed? Wrong us, shall we not revenge?" (Merchant of Venice II i 56-63, paraphrase) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message