Path: utzoo!attcan!uunet!seismo!sundc!pitstop!sun!amdcad!ames!mailrus!tut.cis.ohio-state.edu!rutgers!bellcore!faline!thumper!ulysses!andante!alice!debra
From: debra@alice.UUCP (Paul De Bra)
Newsgroups: comp.unix.microport
Subject: Re: Hard Disks for AT
Message-ID: <8203@alice.UUCP>
Date: 15 Sep 88 14:30:28 GMT
References: <409@ucrmath.UUCP> <1181@umbc3.UMD.EDU> <8198@alice.UUCP> <406@dalek.UUCP>
Reply-To: debra@alice.UUCP ()
Organization: University of Antwerp
Lines: 34

Seth J. Bradley writes:
>I disagree.  I have a system with a Seagate ST4051 and a ST251
>(not ST251-1).  I just ran coretest on the two drives.  Coretest
>is a DOS based drive performance test program that treats the
>drives as raw devices.  It reads in a large block of data and
>gives performance data.  Here is the results of the test I just
>ran:
>
>Drive 	Average Access	Transfer Rate 	Overall Performance
>
>ST4051	38.1 ms		160.6 KB/Sec	2.398
>
>ST251	38.4 ms		162.6 KB/Sec	2.399

This is interesting, cause I have a system with ST4051 and ST251-1
(yes i checked the drive to make sure it is the "-1" version)
I have 2 tests: the first creates a 4Mbyte file, the second reads
512-byte blocks randomly from this file.
The ST4051 and ST251-1 need almost exactly the same time for both
tests. Reading 8192 blocks takes about 330 seconds, which means
40ms per access. I tested an ST4096 too, and it only needed 30ms per
access.

There are 2 possibilities here:
1) I have been sold an ST251 with an ST251-1 label, and so have several
   of my friends (I tested several machines, same result).
2) For average access with Xenix something in the ST251-1 compensates
   the otherwise faster access time. Xenix is not the problem in general
   cause the ST4096 does respond faster.
Sorry I could not verify the behaviour of these drives with Microport
Unix too. The old version I once tried was almost 2 times slower in both
read and write. I assume newer versions will be much faster by now???

Paul.