From owner-freebsd-hackers Wed Apr 9 23:57:03 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id XAA19748 for hackers-outgoing; Wed, 9 Apr 1997 23:57:03 -0700 (PDT) Received: from papillon.lemis.de ([203.239.92.5]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id XAA19716; Wed, 9 Apr 1997 23:56:29 -0700 (PDT) Received: (grog@localhost) by papillon.lemis.de (8.8.4/8.6.12) id PAA01255; Thu, 10 Apr 1997 15:23:13 +0900 (KST) From: grog@lemis.de Message-Id: <199704100623.PAA01255@papillon.lemis.de> Subject: Informix efficiency--any ideas? To: hackers@FreeBSD.ORG (FreeBSD Hackers) Date: Thu, 10 Apr 1997 15:23:05 +0900 (KST) Cc: chat@FreeBSD.ORG (FreeBSD Chat) Reply-To: chat@FreeBSD.ORG (FreeBSD Chat) Organisation: LEMIS, Schellnhausen 2, 36325 Feldatal, Germany Phone: +49-6637-919123 Fax: +49-6637-919122 Reply-to: grog@lemis.de (Greg Lehey) WWW-Home-Page: http://www.FreeBSD.org/~grog X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk This message doesn't really have anything to do with FreeBSD, but possibly somebody here has an idea. As a result, though it's a technical subject, please follow up to -chat. I'm currently in Korea looking at a performance problem with Informix on a MIPS System V platform. We are comparing two MIPS platforms with different System V.4 implementations, A and B. The A machine is an SMP with 2 R4400s at 200 MHz; the B machine is an SMP with 2 R4400s at 150 MHz. B runs a specific, and relatively simple, DB insert program almost exactly twice as fast as A. We did some examination and found that the port of Informix on A performs 21729 system calls to insert 5000 records into a table. Of these, 21403 calls are to semsys. B performs approximately half this number of calls, equally split (what a surprise) between send() and recv(): although it's System V.4, it has a native sockets implementation. I'm obviously following this up with Informix and vendor A, but I'd be interested if anybody here had a view with a different bias. Do semaphores have to be so much less efficient, or could this just be a poor semxxx() implementation on A? Greg