From owner-freebsd-questions Fri Sep 1 23:17:27 2000 Delivered-To: freebsd-questions@freebsd.org Received: from mx1.FreeBSD.org (mx1.FreeBSD.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 058F637B422 for ; Fri, 1 Sep 2000 23:17:26 -0700 (PDT) Received: from thor.afnetinc.com (thor.afnetinc.com [206.40.232.1]) by mx1.FreeBSD.org (Postfix) with ESMTP id 789CD6E26DE for ; Fri, 1 Sep 2000 23:17:25 -0700 (PDT) Received: from 206-40-232-215-pm3-0.afnetinc.com ([206.40.232.215]) by thor.afnetinc.com with smtp (Exim 2.05 #1) id 13V6ar-0003nM-00 for freebsd-questions@FreeBSD.org; Sat, 2 Sep 2000 00:15:57 -0600 From: efinley@efinley.com (Elliot Finley) To: freebsd-questions@FreeBSD.org Subject: FreeBSD 4.1-Stable and Thundering Herds Date: Sat, 02 Sep 2000 06:15:42 GMT Organization: Hiawatha Coal Company Reply-To: efinley@efinley.com Message-ID: <39b1992e.183887987@mail.afnetinc.com> X-Mailer: Forte Agent 1.5/32.451 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: quoted-printable Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Hello, I was wondering if the "Thundering Herd" (http://www.citi.umich.edu/projects/linux-scalability/reports/accept.html= ) problem exists in FreeBSD 4.1-Stable? I have a bunch of processes doing an accept on the same fifo at the same time. I tried serializing the accepts using flock, and it works most of the time, but it seems that I hit a race condition once in a great while when trying to get a lock. I tried the same code without serializing the accepts, and it works, but I was just wondering how well it will scale. Thus the 'Thundering Herd' question. Any pointers/RTFM/anything welcome. P.S. The code is written in Perl 5.005_03 if that matters. --=20 Elliot Finley (efinley@efinley.com) Weird Science! To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message