home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!crdgw1!rdsunx.crd.ge.com!ariel!davidsen
- From: davidsen@ariel.crd.GE.COM (william E Davidsen)
- Newsgroups: comp.unix.pc-clone.32bit
- Subject: Re: Question about ISA/IDE disk throughput (pc newbie)
- Keywords: speed bus IDE EISA ISA SCSI VESA
- Message-ID: <1992Dec22.200930.8089@crd.ge.com>
- Date: 22 Dec 92 20:09:30 GMT
- References: <MUTS.92Dec21193512@PMCS.estec.esa.nl> <1992Dec21.220009.12403@utagraph.uta.edu>
- Sender: usenet@crd.ge.com (Required for NNTP)
- Reply-To: davidsen@crd.ge.com (bill davidsen)
- Organization: GE Corporate R&D Center, Schenectady NY
- Lines: 23
- Nntp-Posting-Host: ariel.crd.ge.com
-
- In article <1992Dec21.220009.12403@utagraph.uta.edu>, turbo@cse.uta.edu (Chris Turbeville) writes:
-
- | 200kB/s sounds about right for IDE (save the newer caching controllers)
- | and it is terribly slow. 486dx's make no improvement on disk transfer
- | speed. It is your 16bit 8Mhz ISA bus which is limiting it. I don't
- | think full blown vmunix would live well in such an environment.
-
- Oops! Better backspace and reread that one in low density... assuming
- a worst case transfer time of six clocks per transfer, you get 1.33M
- transfers/sec, of 2 bytes each, or over 2.6M/sec. A Reasonable
- controller can up that to 4M/sec, but no matter how you slice it the bus
- speed is at least an order of magnitude faster than the rate quoted.
-
- I see transfer rates of 800-1200kB to a controller on an ISA bus,
- oddly enough about the same as I get using the same SCSI disk on a Sun.
-
- I you posit that many IDE controllers and disks have poor transfer
- rates, I would agree, but the supposition about the ISA bus fails on
- both theoretical grounds and by experience.
-
- --
- bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345
- Keyboard controller has been disabled, press F1 to continue.
-