From owner-freebsd-hackers Wed Feb 13 2:37:56 2002 Delivered-To: freebsd-hackers@freebsd.org Received: from mx01-a.netapp.com (mx01-a.netapp.com [198.95.226.53]) by hub.freebsd.org (Postfix) with ESMTP id 8BEA037B404 for ; Wed, 13 Feb 2002 02:37:53 -0800 (PST) Received: from frejya.corp.netapp.com (frejya [10.10.20.91]) by mx01-a.netapp.com (8.11.1/8.11.1/NTAP-1.2) with ESMTP id g1DATV300730; Wed, 13 Feb 2002 02:29:31 -0800 (PST) Received: from cranford-be.eng (localhost [127.0.0.1]) by frejya.corp.netapp.com (8.12.2/8.12.2/NTAP-1.4) with ESMTP id g1DATVJX022252; Wed, 13 Feb 2002 02:29:31 -0800 (PST) Received: from localhost (kmacy@localhost) by cranford-be.eng (8.10.2+Sun/8.10.2) with ESMTP id g1DATUR19315; Wed, 13 Feb 2002 02:29:30 -0800 (PST) Date: Wed, 13 Feb 2002 02:29:30 -0800 (PST) From: Kip Macy To: Alexey Dokuchaev Cc: thttpd@acme.com, hackers@FreeBSD.ORG Subject: Re: THTTPD web server: problems with KQUEUE on FreeBSD 4.5-STABLE In-Reply-To: <20020213160014.A97359@cytherea.weblab.nsu.ru> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII 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 still wonder, whether this problem occurs because of how thttpd does > things, or how FreeBSD implements kqueue stuff, however, I am not sure I take it you don't have any logs for 4.4? > And one more question: out of kqread()/poll()/select() methods, which one > is more likely to perform better, both under normal server access, and Under normal load it doesn't matter. Under heavy load there can be no comparison. see: http://www.kegel.com/dkftpbench/Poller_bench.html This excerpt is the FreeBSD relevant portion: With 1 active socket amongst 100, 1000, or 10000 total sockets, waitAndDispatchEvents takes the following amount of wall-clock time, in microseconds (lower is faster): On a single processor 600Mhz Pentium-III with 512MB of memory, running FreeBSD 4.x-STABLE (results contributed by Jonathan Lemon): pipes 100 1000 10000 30000 select 54 - - - poll 50 552 11559 35178 kqueue 8 8 8 8 (Note: Jonathan also varied the number of active pipes, and found that kqueue's time scaled linearly with that number, whereas poll's time scaled linearly with number of total pipes.) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message