home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.unix.sysv386
- Path: sparky!uunet!maxed!ed
- From: ed@maxed.amg.com (Ed Whittemore)
- Subject: Re: SCSI vs fast SCSI
- Organization: American Micro Group, Ft. Lee NJ
- Date: Sun, 23 Aug 1992 08:53:14 GMT
- Message-ID: <BtFI0s.5Fw@maxed.amg.com>
- References: <1992Aug21.184852.22602@sherpa.uucp>
- Lines: 30
-
- In article <1992Aug21.184852.22602@sherpa.uucp> rac@sherpa.uucp (Roger Cornelius) writes:
- >I posted this to comp.periphs.scsi and received zero replies, so
- >now I'm trying here.
- >
- >
- >In the case of hard disk drives, does fast SCSI provide a significant
- >performance increase over plain SCSI? One controller vendor suggested
- >that fast SCSI pushes the envelope and is more succeptible to failure
- >and for our purposes will provide no benefit. Their controller
- >supports fast SCSI, but they claim plain SCSI will perform just as
- >well for our application (Unix server with disk attached to intelligent
- >caching controller).
- >
- >Can anyone comment?
- >
- >Thanks.
- >--
- >Roger Cornelius rac@sherpa.UUCP ...!uunet!sherpa!rac
-
- There are both faster and not-faster 10 MHz SCSI bus bandwidth (fast
- SCSI) drives. The trick is that the internal transfer rate of the drive
- must take a jump as well for the speed to show up in real thruput
- improvements. However, for truly fast SCSI2-F drives such as the
- HP 2247 and 3010, which have internal rates approaching 40 MHz, you
- see an actual doubling of the speed of copies thru the filesystem,
- over say, a Seagate ST 4767N, a Maxtor P0 or P1, or other 5 MHz
- bus drives with 20 to 30 MHz internal rates.
-
- Ed Whittemore uupsi!maxed!ed ed@maxed.amg.com
- American Micro Group, Inc. 201-944-3293
-