home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!sun-barr!cs.utexas.edu!qt.cs.utexas.edu!yale.edu!ira.uka.de!ira.uka.de!math.fu-berlin.de!unidui!Germany.EU.net!mcsun!sunic!seunet!kullmar!piraya!overdose!Roger_Nordin
- From: Roger_Nordin@atb.bbs.bad.se (Roger Nordin)
- Newsgroups: comp.periphs.scsi
- Subject: Re: 1 GB 3.5" SCSI disks
- Message-ID: <OD.6badnetOA92-901-302p0_53b52437@piraya.bad.se>
- Date: Wed, 18 Nov 92 21:56:17 +0000
- References: <1992Nov11.175733.15236@b11.b11.ingr.com>
- Sender: BadNet@piraya.bad.se
- Reply-To: Roger_Nordin@atb.bbs.bad.se (Roger Nordin)
- Distribution: world
- Organization: ANet Test Bench, Karlstad, Sweden
- OD-Comment-To: Internet_Gateway
- Lines: 36
-
- > >Testing with a 262144 byte, MEMF_FAST, LONG-aligned buffer.
- > >Create file: 940863 bytes/sec | CPU Available: 65%
- > >Write to file: 1699686 bytes/sec | CPU Available: 75%
- > >Read from file: 3195287 bytes/sec | CPU Available: 64%
- > >
- > >The same test on a partition starting in the middle of the drive:
- > >
- > >Testing with a 262144 byte, MEMF_FAST, LONG-aligned buffer.
- > >Create file: 688128 bytes/sec | CPU Available: 71%
- > >Write to file: 1081662 bytes/sec | CPU Available: 75%
- > >Read from file: 2244160 bytes/sec | CPU Available: 59%
- >
- > This is curious. The DEC DSP3105 is not a "zoned" disk drive. The
- > data rate is constant regardless of where you are on the disk.
- > The sustained transfer rate (57 sec/track * 512 bytes per sector * 90rps)
- > is 2.6MB/sec. The 2.2MB/sec you see is reasonable. The 3.1 is not. I
- > suspect disk or system cache is responsible for the "inflated"
- > value.
-
- If a cache had been involved, would the "CPU Available" not have showed lower
- figures? Software caches always eats a lot CPU time when reaching from the
- cache, since that is a 100% dependant CPU operation itself.
-
- I don't have the data for the DSP3105 disk (actually, I just popped in here),
- but if the subject line is still correct, then we are talking about 1Gb sized
- harddisks. Now, 1Gb 3.5" without zone recording?! Why waste so much disk space?
- Wouldn't it make the disk more unreliable, if it's not adapting the data
- stored on each cylinder to the size (and thus density) of the cylinder itself?
- You'd be pushing the inner cylinders to the limits, while the outermost would
- not be used even to half their capacity.
-
- _ . __
- Roger Nordin _ // /!Net /est Bench
- 2:203/602.0@fidonet \X/ Internet: Roger_Nordin@bad.se
-
- ---
-