From owner-freebsd-hardware Tue Sep 30 11:03:12 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id LAA10425 for hardware-outgoing; Tue, 30 Sep 1997 11:03:12 -0700 (PDT) Received: from peeper.my.domain ([208.128.8.69]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id LAA10420 for ; Tue, 30 Sep 1997 11:03:03 -0700 (PDT) Received: (from tom@localhost) by peeper.my.domain (8.8.7/8.7.3) id MAA17795; Tue, 30 Sep 1997 12:52:51 -0500 (CDT) Message-ID: <19970930125250.01095@my.domain> Date: Tue, 30 Sep 1997 12:52:50 -0500 From: Tom Jackson To: freebsd-hardware@FreeBSD.ORG Subject: Re: Opinions please Was: Micropolis Runs Too HOT! References: <199709132341.NAA23393@pegasus.com> <19970920131634.08085@my.domain> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.81e In-Reply-To: <19970920131634.08085@my.domain>; from Tom Jackson on Sat, Sep 20, 1997 at 01:16:34PM -0500 Sender: owner-freebsd-hardware@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Sat, Sep 20, 1997 at 01:16:34PM -0500, Tom Jackson wrote: > In my previous post I disCussed the heating situation with this 7200rpm scsi > Micropolis drive. I now have two 4-1/2in cooling fans on the midi case, one > blowing over the Mic located in two 5-1/4 in bays and one sucking out at the > back of the case. Although I have return approval for a Fujitsu M2954S, many > claim it will run just hot as the Micropolis. I have run iozone on all of my > drives trying to decide if I should exchange. Although the Mic seems to write > at a good rate the read specs don't look good at all. Would anyone with good > knowledge of this please comment. I need to decide soon. > > iozone single results: > --------------------------------------------------------- > In the three runs, done separately, the disk cache sized to 11 megabytes. > sd0 is a Micropolis 4421 2.1gb 5400rpm scsi, sd1 is a Micropolis 4345 4.5gb > 7200rpm scsi, and sd2 is a Fujitsu M1606sau 1.08gb 5400rpm scsi. > > IOZONE: Performance Test of Sequential File I/O -- V2.01 (10/21/94) > By Bill Norcott > > Operating System: FreeBSD 2.x -- using fsync() > > Send comments to: b_norcott@xway.com > > IOZONE writes a 44 Megabyte sequential file consisting of > 5632 records which are each 8192 bytes in length. > It then reads the file. It prints the bytes-per-second > rate at which the computer can read and write files. > > > Writing the 44 Megabyte file, '/usr/ports/iozone.tmp'...11.265625 seconds > Reading the file...7.914062 seconds on sd0 > > IOZONE performance measurements: > 4095409 bytes/second for writing the file > 5829792 bytes/second for reading the file > <--- > > Writing the 44 Megabyte file, '/usr/iozone.tmp'...6.625000 seconds > Reading the file...6.257812 seconds on sd1 > > IOZONE performance measurements: > 6964127 bytes/second for writing the file > 7372759 bytes/second for reading the file > <--- > > Writing the 44 Megabyte file, '/mnt1/iozone.tmp'...6.335938 seconds > Reading the file...5.953125 seconds on sd2 > > IOZONE performance measurements: > 7281849 bytes/second for writing the file > 7750105 bytes/second for reading the file > ----------------------------------------------------------- > iozone concurrent results: > > Concurrent run to all three drives. Disk cache sized itself to 11mb. > > IOZONE: Performance Test of Sequential File I/O -- V2.01 (10/21/94) > By Bill Norcott > > Operating System: FreeBSD 2.x -- using fsync() > > Send comments to: b_norcott@xway.com > > IOZONE writes a 44 Megabyte sequential file consisting of > 5632 records which are each 8192 bytes in length. > It then reads the file. It prints the bytes-per-second > rate at which the computer can read and write files. > > > Writing the 44 Megabyte file, '/usr/ports/iozone.tmp'...11.250000 seconds > Reading the file...7.453125 seconds on sd0 (MC4421) > > IOZONE performance measurements: > 4101097 bytes/second for writing the file > 6190335 bytes/second for reading the file > <--- > > Writing the 44 Megabyte file, '/usr/iozone.tmp'...6.585938 seconds > Reading the file...6.304688 seconds on sd1 (MC4345NS) > > IOZONE performance measurements: > 7005433 bytes/second for writing the file > 7317943 bytes/second for reading the file > <--- > > Writing the 44 Megabyte file, '/mnt1/iozone.tmp'...6.359375 seconds > Reading the file...5.867188 seconds on sd2 (Fuj M1606S) > > IOZONE performance measurements: > 7255012 bytes/second for writing the file > 7863621 bytes/second for reading the file > > > I can't tell if this bad performance or the disk might be damaged > > Tom > --- > > Well no one commented but I did exchange the drive for a Fujitsu M2954S. Here's the iozone for it: Writing the 44 Megabyte file, '/usr/iozone.tmp'...6.078125 seconds Reading the file...7.281250 seconds IOZONE performance measurements: for sd1 (Fujitsu M2954S) 7590719 bytes/second for writing the file 6336459 bytes/second for reading the file As you can see, much much better! Boy am I glad I swapped. Tom