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: Supporting *BIG* Drives <was: Adaptec 1742A on ISC-3.0>
- Message-ID: <1992Dec22.194754.7583@crd.ge.com>
- Date: 22 Dec 92 19:47:54 GMT
- References: <1992Dec19.004318.10508@twg.com>
- 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: 27
- Nntp-Posting-Host: ariel.crd.ge.com
-
- In article <1992Dec19.004318.10508@twg.com>, david@twg.com (David Herron) writes:
-
- | Our drive is "only" 600 megs and it's working fine and without problems
- | even. And we're mucho happy with the performance on this new system. Going
- | from a 16 MHz '386 (8 megs RAM and 600 megs ESDI disk) to 50 MHz '486
- | on EISA (16 megs RAM and 600 megs SCSI) reduced the time to compile
- | our product from 10 hours(!) to 1.5 hours...
- |
- | Dunno if things is any worse with hughmungo drives like 2 gigs.
- |
- | I'm still running the SCSI card in 154x mode with the above settings.
- | I tried configuring it for enhanced mode, but the system refused to
- | boot then. Don't remember the problem but was along the lines of being
- | unable to find the disk. And after 1.5 weeks "wasted" in configuring
- | this system I'm tired of fiddling with it...
-
- I'm convinced that the secret to good compilation performance is using
- a cache with writeback enabled. failing in this add about 16MB of RAM
- for the system to use for buffers, use the shared libraries, and put
- the compiler on ramdisk.
-
- Use sar to watch the "queue time" of i/o requests, and add memory
- until the time stops dropping like a rock.
-
- --
- bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345
- Keyboard controller has been disabled, press F1 to continue.
-