From owner-freebsd-questions Fri Aug 15 21:32:39 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id VAA06486 for questions-outgoing; Fri, 15 Aug 1997 21:32:39 -0700 (PDT) Received: from fly.HiWAAY.net (root@fly.HiWAAY.net [208.147.154.56]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id VAA06476 for ; Fri, 15 Aug 1997 21:32:30 -0700 (PDT) Received: from nospam.hiwaay.net (tnt1-201.HiWAAY.net [208.147.147.201]) by fly.HiWAAY.net (8.8.6/8.8.6) with ESMTP id XAA26983 for ; Fri, 15 Aug 1997 23:32:12 -0500 (CDT) Received: from localhost (localhost [127.0.0.1]) by nospam.hiwaay.net (8.8.6/8.8.4) with ESMTP id WAA05741 for ; Fri, 15 Aug 1997 22:57:37 -0500 (CDT) Message-Id: <199708160357.WAA05741@nospam.hiwaay.net> X-Mailer: exmh version 2.0zeta 7/24/97 To: questions@FreeBSD.ORG From: dkelly@hiwaay.net Subject: Re: FreeBSD is slower than Linux !? In-reply-to: Message from Paul Dekkers of "Fri, 15 Aug 1997 11:56:00 +0200." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 15 Aug 1997 22:57:37 -0500 Sender: owner-freebsd-questions@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Paul Dekkers writes: > > And, by the way, when using an asynced disk, it isn't absolutely unsafe, > am I right? When running sync once a day it writes everything to disk or Sync once a day? Forgot the name of the daemon Linux runs that syncs every 30 seconds. > not ?! (maybe mount /var with sync, en /www (or smth) with async ...) > > nic% time > 0.080u 0.070s 0:06.48 2.3% 176+96k 0+0io 0pf+0w > nic% time dd if=/dev/zero of=test bs=1024 count=5000 > 5000+0 records in > 5000+0 records out > 0.010u 1.010s 0:01.83 55.7% 0+0k 0+0io 13pf+0w > nic% time sync > 0.010u 0.320s 0:01.97 16.7% 11+0k 0+0io 8pf+0w The above is a Linux system, right? And with "time sync" you are demonstrating that data is still cached and not yet written? FreeBSD dd gives a data rate: nospam: {306} time dd if=/dev/zero of=test bs=1024 count=5000 5000+0 records in 5000+0 records out 5120000 bytes transferred in 1.076674 secs (4755386 bytes/sec) 0.1u 0.9s 0:01.34 76.1% 82+229k 5+81io 5pf+0w nospam: {307} time dd if=/dev/zero of=test bs=50k count=100 100+0 records in 100+0 records out 5120000 bytes transferred in 0.571046 secs (8966002 bytes/sec) 0.0u 0.5s 0:00.73 79.4% 83+349k 1+95io 0pf+0w nospam: {308} time sync 0.0u 0.0s 0:00.01 200.0% 116+262k 2+34io 1pf+0w That's pretty good too, 200% CPU utilization on a single CPU? nospam: {309} ls -l test -rw-r--r-- 1 dkelly dkelly 5120000 Aug 15 22:41 test nospam: {310} mount /dev/sd0a on / (local) /dev/sd0s2f on /usr (NFS exported, local) /dev/sd0s2e on /var (local) procfs on /proc (local) nospam: {311} pwd /usr/home/dkelly/rc5v2 nospam: {312} uname -a FreeBSD nospam.hiwaay.net 2.2-STABLE FreeBSD 2.2-STABLE #0: Wed Aug 13 21:45:26 CDT 1997 dkelly@nospam.hiwaay.net:/usr/src/sys/compile/NEXGEN i386 nospam: {313} Hey! What gives? This pokey system which does "make world" in 9 hours beat the other results that were posted. Maybe I'll quit lusting for a Tomcat IV MB and either Cyrix 6x86MX-PR200 or Pentium 166 w/MMX, but still tempted by Pentium Pro's. Whatever I get, *this* time I'll get an FPU. -- David Kelly N4HHE, dkelly@hiwaay.net ===================================================================== The human mind ordinarily operates at only ten percent of its capacity -- the rest is overhead for the operating system.