*BSD News Article 7464


Return to BSD News archive

Xref: sserve comp.arch:27896 comp.unix.bsd:7513 comp.os.linux:14883
Path: sserve!manuel.anu.edu.au!munnari.oz.au!spool.mu.edu!agate!doc.ic.ac.uk!uknet!cf-cm!news
From: spxpmf@thor.cf.ac.uk (Phillip Fayers)
Newsgroups: comp.arch,comp.unix.bsd,comp.os.linux
Subject: Re: IDE faster than Mips SCSI disk
Message-ID: <13481.9211061131@thor.cf.ac.uk>
Date: 6 Nov 92 11:31:23 GMT
References: <1992Nov6.033942.21194@ntuix.ntu.ac.sg>
Sender: news@cm.cf.ac.uk (Network News System)
Organization: Physics Dept. University of Wales, Cardiff
Lines: 52
X-Mailer: Cardiff Computing Maths PP Mail Open News Gateway

In article <1992Nov6.033942.21194@ntuix.ntu.ac.sg> eoahmad@ntuix.ntu.ac.sg (Othman Ahmad) writes:
|These figures are not meant to be a thorough study. More of a provocation
|for more soul searching. The widespread belief that SCSI-2 is "defininely
|faster than IDE" is questionable, in fact under some circumstances, entirely
|false.
|	The results are not surprising for those who study the technical
|specs of these disks. IDE has more potential for high speed because of its
|lack of protocaol overhead. Its closeness to the disk also make it very
|reliable and efficient.
|	There are advanced disks which can do simultaneour multiple-head reads,
|but these techniques can also be used for IDE as well.
|	IDE is just a simple interface definition, just like SCSI-2, but IDE,
|is optimised for HARD DISKS, SCSI is not. SCSI is more general purpose.
|
|The mips machine under test runs on ultrix. Although it has up to 10Gbyte 
|of hard-disk, it is not so heavily loaded. We only use it for email and
|news feed. Fragmentation can be severe because I cannot even have 32 
|megabyte free space in /usr/tmp , only 30 Mbytes.
|
|Writing the 30 Megabyte file, 'iozone.tmp'...91.000000 seconds
|Reading the file...49.000000 seconds
|
|IOZONE performance measurements:
|	345684 bytes/second for writing the file
|	641985 bytes/second for reading the file
|****************************************************************************
|This PC machine runs on 386bsd. Using 1 megabyte test file it is faster than
|a similar 486/50Mhz EISA SCSI-2 hard-disk.
|
|486/33 Maxtor 7120 200Mbyte
|
|Writing the 32 Megabyte file, 'iozone.tmp'...91.283333 seconds
|Reading the file...67.116667 seconds
|
|IOZONE performance measurements:
|	367586 bytes/second for writing the file
|	499942 bytes/second for reading the file

I wouldn't put this difference down to SCSI versus IDE, I would put most
of it down to the poor IO performance of Ultrix. We recently had a DEC
announcement of Ultrix 4.3 where they claim to have improved file IO by
100-300%, a figure I'm inclined to believe. I ran the Byte UNIX
benchmarks on a DECsystem 8300 that we have here, its file IO
performance was comparable to that of a Sun SPARCstation IPC's internal
(slow) 207 MByte drive. File copy on both systems was approx 280 kBytes
per second. In comparison a SUN 4/360 we have here turned in 1440 kBytes
per second on file copy, this was using a 1.4 GByte SCSI disk on a slow
SCSI II interface (5 Mbytes per second max).
-- 

Phillip Fayers                      email: fayers@cardiff.ac.uk
Sun admin/support/programming       phone: 0222 874000 x 5282 (UK)