home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.apollo
- Path: sparky!uunet!usc!sdd.hp.com!caen!destroyer!terminator!pisa.citi.umich.edu!rees
- From: rees@pisa.citi.umich.edu (Jim Rees)
- Subject: Re: DN10K's for sale
- Message-ID: <5ad0b55b.1bc5b@pisa.citi.umich.edu>
- Sender: news@terminator.cc.umich.edu (Usenet Owner)
- Reply-To: Jim.Rees@umich.edu
- Organization: University of Michigan IFS Project
- References: <9208271756.AA07955@gotham.hwcae.az.Honeywell.COM>
- Date: Thu, 27 Aug 1992 19:13:48 GMT
- Lines: 15
-
- In article <9208271756.AA07955@gotham.hwcae.az.Honeywell.COM>, davidy@HWCAE.AZ.HONEYWELL.COM (David Young -davidy K26E2 x2968) writes:
-
- min:secs
- 0:13.84 copy disk-to-same-disk //indy --> //indy
- 1:20.73 copy disk-to-same-disk //gotham --> //gotham
-
- //indy is a 4 CPU 512MB RAM 2 ESDI disk (sector striped) DN10k
- //gotham is a 32MB RAM 2 SCSI disk (concat striped) 425t
-
- The data copied was a 34MB Mentor/GeoDraw schematic database. Now
- that's what I call I/O performance (the DN10k).
-
- No, that's what you call write-behind. Far more interesting would be a
- comparison with a 16 Mb file (or a 16 Gb file). Don't forget to flush vm
- first.
-