From owner-freebsd-smp Mon Feb 28 7:31: 8 2000 Delivered-To: freebsd-smp@freebsd.org Received: from mailbox.reptiles.org (mailbox.reptiles.org [198.96.117.155]) by hub.freebsd.org (Postfix) with ESMTP id 7AB5737B50E for ; Mon, 28 Feb 2000 07:31:05 -0800 (PST) (envelope-from jim@reptiles.org) Received: from localhost (1424 bytes) by mailbox.reptiles.org via sendmail with P:stdio/R:bind_hosts/T:inet_zone_bind_smtp (sender: ) (ident using unix) id for ; Mon, 28 Feb 2000 10:30:57 -0500 (EST) (Smail-3.2.0.108 1999-Sep-19 #3 built 1999-Oct-27) Date: Mon, 28 Feb 2000 10:30:57 -0500 From: Jim Mercer To: freebsd-smp@freebsd.org Subject: current working SMP mboards? Message-ID: <20000228103056.I606@reptiles.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0i Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org i poked about on the freebsd website, and found that the information about SMP appears to be somewhat stale. i'm not on freebsd-smp, so direct replies would be appreciated. i'm running a fairly large/busy postgresql database. i'm considering moving it to a multi-processor (start with 2, maybe add more) machine. i'm considering the following for 3.4-STABLE (or 4.0-RELEASE): ASUS (P2B-DS or P2L97-DS) 512M RAM 1 x scsi drive for OS/applications 5 x 9/18 gig U2W drives under vinum raid5 for data my understanding of the SMP stuff is that if the application is not multi-threaded, then SMP doesn't do much. my theory was that with a dual processor, each incantation of the application would fire up on alternating processors. is this correct? or is SMP effectively useless unless my application (primarily postgresql) is multi-threaded. -- [ Jim Mercer jim@reptiles.org +1 416 506-0654 ] [ Reptilian Research -- Longer Life through Colder Blood ] [ Don't be fooled by cheap Finnish imitations; BSD is the One True Code. ] To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Feb 28 8: 8:27 2000 Delivered-To: freebsd-smp@freebsd.org Received: from www.a-znet.com (www.a-znet.com [209.177.29.5]) by hub.freebsd.org (Postfix) with ESMTP id D4C1237B87C for ; Mon, 28 Feb 2000 08:08:23 -0800 (PST) (envelope-from klmac@a-znet.com) Received: from klmac.a-znet.com (client22.a-znet.com [209.177.29.22]) by www.a-znet.com (8.8.7/8.8.7) with ESMTP id LAA15180 for ; Mon, 28 Feb 2000 11:29:46 -0500 Message-Id: <4.3.0.20000228100851.00aaf880@pop-server.twcny.rr.com> X-Sender: klmac@mail.a-znet.com X-Mailer: QUALCOMM Windows Eudora Version 4.3 Date: Mon, 28 Feb 2000 10:08:58 -0500 To: freebsd-smp@freebsd.org From: Ken McKittrick Subject: current working SMP mboards? Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org i poked about on the freebsd website, and found that the information about SMP appears to be somewhat stale. i'm not on freebsd-smp, so direct replies would be appreciated. i'm running a fairly large/busy postgresql database. i'm considering moving it to a multi-processor (start with 2, maybe add more) machine. i'm considering the following for 3.4-STABLE (or 4.0-RELEASE): ASUS (P2B-DS or P2L97-DS) 512M RAM 1 x scsi drive for OS/applications 5 x 9/18 gig U2W drives under vinum raid5 for data my understanding of the SMP stuff is that if the application is not multi-threaded, then SMP doesn't do much. my theory was that with a dual processor, each incantation of the application would fire up on alternating processors. is this correct? or is SMP effectively useless unless my application (primarily postgresql) is multi-threaded. -- [ Jim Mercer jim@reptiles.org +1 416 506-0654 ] [ Reptilian Research -- Longer Life through Colder Blood ] [ Don't be fooled by cheap Finnish imitations; BSD is the One True Code. ] To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message Ken McKittrick Network Engineer A-Znet.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Feb 28 8:17:11 2000 Delivered-To: freebsd-smp@freebsd.org Received: from kronos.alcnet.com (kronos.alcnet.com [63.69.28.22]) by hub.freebsd.org (Postfix) with ESMTP id 8848E37BA2C for ; Mon, 28 Feb 2000 08:17:07 -0800 (PST) (envelope-from kbyanc@posi.net) X-Provider: ALC Communications, Inc. http://www.alcnet.com/ Received: from localhost (kbyanc@localhost) by kronos.alcnet.com (8.9.3/8.9.3/antispam) with ESMTP id LAA17188; Mon, 28 Feb 2000 11:17:01 -0500 (EST) Date: Mon, 28 Feb 2000 11:17:01 -0500 (EST) From: Kelly Yancey X-Sender: kbyanc@kronos.alcnet.com To: Jim Mercer Cc: freebsd-smp@FreeBSD.ORG Subject: Re: current working SMP mboards? In-Reply-To: <20000228103056.I606@reptiles.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Mon, 28 Feb 2000, Jim Mercer wrote: > i'm considering the following for 3.4-STABLE (or 4.0-RELEASE): > ASUS (P2B-DS or P2L97-DS) > 512M RAM > 1 x scsi drive for OS/applications > 5 x 9/18 gig U2W drives under vinum raid5 for data I cannot personally comment on your hardware setup, but I recall others on -smp indicating success with the ASUS SMP motherboards. > > my understanding of the SMP stuff is that if the application is not > multi-threaded, then SMP doesn't do much. > If the application is multi-threaded, currently SMP doesn't do much. FreeBSD has user-land threads and the smallest kernel schedulable entities are processes, not threads. > my theory was that with a dual processor, each incantation of the application > would fire up on alternating processors. Assuming each application is it's own process, then you would be utilizing both CPUs. > > is this correct? or is SMP effectively useless unless my application > (primarily postgresql) is multi-threaded. > Your scenario should work precisely because, at least under current versions of FreeBSD, SMP is most effective in non-multi-threaded applications. Kelly -- Kelly Yancey - kbyanc@posi.net - Richmond, VA Analyst / E-business Development, Bell Industries http://www.bellind.com/ Maintainer, BSD Driver Database http://www.posi.net/freebsd/drivers/ Coordinator, Team FreeBSD http://www.posi.net/freebsd/Team-FreeBSD/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Feb 28 8:21:48 2000 Delivered-To: freebsd-smp@freebsd.org Received: from mailbox.reptiles.org (mailbox.reptiles.org [198.96.117.155]) by hub.freebsd.org (Postfix) with ESMTP id 2461C37B8A5 for ; Mon, 28 Feb 2000 08:21:44 -0800 (PST) (envelope-from jim@reptiles.org) Received: from localhost (3633 bytes) by mailbox.reptiles.org via sendmail with P:stdio/R:bind_hosts/T:inet_zone_bind_smtp (sender: ) (ident using unix) id for ; Mon, 28 Feb 2000 11:21:40 -0500 (EST) (Smail-3.2.0.108 1999-Sep-19 #3 built 1999-Oct-27) Date: Mon, 28 Feb 2000 11:21:40 -0500 From: Jim Mercer To: Bart van Leeuwen Cc: freebsd-smp@freebsd.org Subject: Re: current working SMP mboards? Message-ID: <20000228112140.L606@reptiles.org> References: <20000228103056.I606@reptiles.org> <38BA9C6D.70769CD@ixori.demon.nl> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0i In-Reply-To: <38BA9C6D.70769CD@ixori.demon.nl>; from bart@ixori.demon.nl on Mon, Feb 28, 2000 at 05:03:57PM +0100 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Mon, Feb 28, 2000 at 05:03:57PM +0100, Bart van Leeuwen wrote: > > i'm considering the following for 3.4-STABLE (or 4.0-RELEASE): > > ASUS (P2B-DS or P2L97-DS) > > 512M RAM > > 1 x scsi drive for OS/applications > > 5 x 9/18 gig U2W drives under vinum raid5 for data > > First of all, I am using virtually the same config, the p2b-ds works > fine and the scsi stuff works out of the box on both 3.x and 4.0 cool. > > my theory was that with a dual processor, each incantation of the > > application would fire up on alternating processors. > > In quite a few cases that works fine. Don't count on it that the app > will keep running on the same cpu all the time, but when you run some > app twice and they both require cpu time, fbsd will distribute the load > over its 2 cpus. postgresql works with a main process (which coordinates reads/writes to the actual database) and a pairing of client/server processes to actually deal with the query. so, if i fire up a query, there are 3 processes involved, one persistent, two transient. if there are two cpu's, then the processes have more available CPU to get the job done, right? i understand that a single, non-threaded, process will not use cycles from both processes concurrently, but two processes may end up using independent processors for CPU cycles. i gather that the processes might end up on the same CPU, but since i have many queries running at the same time, it would have some advantage to have the processes round-robin'ing the CPU's. > > is this correct? or is SMP effectively useless unless my application > > (primarily postgresql) is multi-threaded. > > Not mostly useless, tho its unlikely to double the performance.. doubling the performance would be way cool, but even a 25% increase in performance would be an acceptable upgrade from where we are now. > Its important to realize that a single instance of an application can > only be run by 2 cpus simultaniously if it is multi threaded. A non > multi threaded app is unlikely to show any performance gain on smp. What > SMP does do in such cases however is make that you can run more on that > machine.. ie, it doesn't get faster, it gets more capacity. (ie, the > capacity to run a 2nd instance of an application without any impact on > the first instance) assuming disk I/O is not a factor (which it may be), if i have a process which is processor intensive, i can run two instances of that process in parallel, right? so if the process takes 30 seconds to complete, then it would be 60 seconds in serial, and 30 seconds (plus something no doubt) in parallel. or are you saying that even when running the processes in parallel, that there are wait-states or something that would make the parallel running time similar to the serial running time? -- [ Jim Mercer jim@reptiles.org +1 416 506-0654 ] [ Reptilian Research -- Longer Life through Colder Blood ] [ Don't be fooled by cheap Finnish imitations; BSD is the One True Code. ] To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Feb 28 9:10:21 2000 Delivered-To: freebsd-smp@freebsd.org Received: from thomson.iqm.unicamp.br (thomson.iqm.unicamp.br [143.106.51.247]) by hub.freebsd.org (Postfix) with ESMTP id 5AD7937B902 for ; Mon, 28 Feb 2000 09:10:12 -0800 (PST) (envelope-from fabio@thomson.iqm.unicamp.br) Received: (from fabio@localhost) by thomson.iqm.unicamp.br (8.9.3/8.9.2) id OAA73898; Mon, 28 Feb 2000 14:08:03 -0300 (EST) (envelope-from fabio) From: Fabio Cesar Gozzo Message-Id: <200002281708.OAA73898@thomson.iqm.unicamp.br> Subject: Re: current working SMP mboards? In-Reply-To: <20000228112140.L606@reptiles.org> from Jim Mercer at "Feb 28, 2000 11:21:40 am" To: jim@reptiles.org (Jim Mercer) Date: Mon, 28 Feb 2000 14:08:03 -0300 (EST) Cc: bart@ixori.demon.nl, freebsd-smp@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL43 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Jim Mercer wrote: >On Mon, Feb 28, 2000 at 05:03:57PM +0100, Bart van Leeuwen wrote: >> Its important to realize that a single instance of an application can >> only be run by 2 cpus simultaniously if it is multi threaded. A non >> multi threaded app is unlikely to show any performance gain on smp. What >> SMP does do in such cases however is make that you can run more on that >> machine.. ie, it doesn't get faster, it gets more capacity. (ie, the >> capacity to run a 2nd instance of an application without any impact on >> the first instance) > >assuming disk I/O is not a factor (which it may be), if i have a process which >is processor intensive, i can run two instances of that process in parallel, >right? so if the process takes 30 seconds to complete, then it would be >60 seconds in serial, and 30 seconds (plus something no doubt) in parallel. > >or are you saying that even when running the processes in parallel, that >there are wait-states or something that would make the parallel running time >similar to the serial running time? If you are looking for experiences, I've been running a parallel calculation program on ASUS P2B-DS (latest 4.0 snapshot) having a gain ranging from 1.8 to 1.95. Of course, that depends on the parallel implementation. My program uses fork()/wait() to spaw another process. YMMV. -- ************************************************** Fabio Gozzo fabio@iqm.unicamp.br State University of Campinas UNICAMP Chemistry Institute http://thomson.iqm.unicamp.br ************************************************** To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Feb 29 2: 5:20 2000 Delivered-To: freebsd-smp@freebsd.org Received: from outmail.utsunomiya-u.ac.jp (outmail.utsunomiya-u.ac.jp [160.12.196.3]) by hub.freebsd.org (Postfix) with ESMTP id 0453E37BAB5; Tue, 29 Feb 2000 02:05:13 -0800 (PST) (envelope-from yokota@zodiac.mech.utsunomiya-u.ac.jp) Received: from zodiac.mech.utsunomiya-u.ac.jp (IDENT:9ygUSFg3w+5QIRfW6mF8O/xy1VdRkZ0m@zodiac.mech.utsunomiya-u.ac.jp [160.12.42.1]) by outmail.utsunomiya-u.ac.jp (8.9.3/3.7Wpl2) with ESMTP id TAA20661; Tue, 29 Feb 2000 19:05:11 +0900 (JST) Received: from zodiac.mech.utsunomiya-u.ac.jp (zodiac.mech.utsunomiya-u.ac.jp [160.12.42.1]) by zodiac.mech.utsunomiya-u.ac.jp (8.7.6+2.6Wbeta7/3.4W/zodiac-May96) with ESMTP id TAA01841; Tue, 29 Feb 2000 19:11:14 +0900 (JST) Message-Id: <200002291011.TAA01841@zodiac.mech.utsunomiya-u.ac.jp> To: freebsd-smp@freebsd.org, freebsd-current@freebsd.org Cc: yokota@zodiac.mech.utsunomiya-u.ac.jp Subject: AP #1 failed! Date: Tue, 29 Feb 2000 19:11:13 +0900 From: Kazutaka YOKOTA Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In the last few days, I occasionally see the following message when I boot the 4.0-CURRENT kernel. AP #1 (PHY# 1) failed! panic y/n [y]? This doesn't always happen. But, roughly once in a few reboots. My MB is Gigabyte GA-6BXD with two Celerons (400MHz). Do I have a hardware problem? How should I diagnose the problem? Kazu To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message