Date: Fri, 11 Sep 1998 11:47:18 -0600 (MDT) From: "Kenneth D. Merry" <ken@plutotech.com> To: dag-erli@ifi.uio.no (Dag-Erling =?iso-8859-1?Q?Co=EFdan?= =?iso-8859-1?Q?Sm=F8rgrav?=) Cc: vallo@matti.ee, freebsd-scsi@FreeBSD.ORG Subject: Re: ncr: frequent ccb errors and crashes Message-ID: <199809111747.LAA20788@panzer.plutotech.com> In-Reply-To: <xzpaf4662vr.fsf@skejdbrimir.ifi.uio.no> from =?ISO-8859-1?Q?Dag=2DErling_Co=EFdan__Sm=F8rgrav?= at "Sep 11, 98 06:16:40 pm"
next in thread | previous in thread | raw e-mail | index | archive | help
Dag-Erling Coïdan Smørgrav wrote... > Vallo Kallaste <vallo@matti.ee> writes: > > Yeah, nice to see that you have IBM disks, but these are truly overpriced > > here, Quantum costs about half of the IBM disks. Do you suggest replacing > > controller or what I can do to test that specific controller. Maybe I'm > > having a faulty controller, that is. > > Bullshit. I don't know about the Baltic countries, but in Norway, > large IBM SCSI drives cost much less than equivalent Quantum or > Seagate drives. Example (from Tech Computers in Oslo): > > UltraStar 9ZXS 9,1 GB DDRS-39110W > 6.3ms, 1024kB, 10000 rpm, UW-SCSI > 6.650,- 8.180,- > > Cheetah 9.1 GB ST19101W > 8ms, 512kB, 10000 rpm, UW-SCSI > 7.290,- 8.967,- > > The IBM drive is not only cheaper, but faster too. (the first price is > without sales tax, the second is with tax) The second drive is a first generation Cheetah. The second generation cheetahs are pretty nice, and I believe they are slightly faster than the IBM Ultrastar 9ZX. Take a look at: http://www.seagate.com/disc/cheetah/cheetah.shtml and: http://www.storage.ibm.com/hardsoft/diskdrdl/ultra/9zxdata.htm Compare the "media data rate" on IBM's page with the "internal transfer rate" on Seagate's page. You're right that the first generation Cheetah is slower than the 9ZX. The second generation Cheetah, however, is faster than the 9ZX. First generation Cheetah: 122 - 177 Mbits/sec UltraStar 9ZX: 129.3 - 204.8 Mbits/sec Second generation Cheetah: 152 - 231 Mbits/sec IIRC, those numbers don't directly translate to MB/sec. You have to do some sort of calculation to get actual MB/sec. I've got an UltraStar 9ZX at home, and it's a very nice drive. Here are some iozone results: ======================================================================== {panzer:/a/ken:103:0} iozone 1024 65536 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 1024 Megabyte sequential file consisting of 16384 records which are each 65536 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 1024 Megabyte file, 'iozone.tmp'...73.773438 seconds Reading the file...64.851562 seconds IOZONE performance measurements: 14554585 bytes/second for writing the file 16556915 bytes/second for reading the file ======================================================================== One interesting thing about the IBM drives is that they have an on-board temperature sensor. The drive will supposedly return transactions with certain sense codes when it gets too hot. Ken -- Kenneth Merry ken@plutotech.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199809111747.LAA20788>