Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!ames!ucsd!rutgers!att!pegasus!psrc From: psrc@pegasus.ATT.COM (Paul S. R. Chisholm) Newsgroups: comp.sys.att Subject: Re: Fast/Slow 6386s Summary: some guesses Keywords: ST506 vs. ESDI IX386 Message-ID: <2624@pegasus.ATT.COM> Date: 26 Feb 89 05:17:01 GMT References: <1060@vsi.COM> Distribution: comp Organization: AT&T Bell Laboratories Lines: 35 <"Would you like me to summon Data so he could offer a few dozen synonyms?"> In article <1060@vsi.COM>, droman@vsi.COM (Dave Droman) writes: > I have a customer who is running two (almost) identical 6386 16Mz systems. > One has a 135MB ESDI drive and the other has a 40MB ST506 drive; both have > 4MB RAM. > > Both the systems are running ix386 and VP/IX and the identical applications. > The 40MB is "blowing the doors off" the 135MB. Indeed, DOS applications under > VP/IX are actually reasonable on the 40. I certainly imagine that, all other things being equal, the EDSI drive should be much faster than the ST-506. The most obvious thing that's *not* equal is the size of the disks. Maybe it's the layout of the file systems? Ideally, you should keep all of your most frequently accessed files (and file systems) near one another on the disk. If the big disk is divided up into lots of large, mostly empty file systems, the head will spend a lot of its time travelling over unused tracks. Unfortunately, I can't think of a good way to measure this; and fixing this involves backing up and restoring most of your hard disk. The other thing to check is how well-cached the disks are. If the 40M system is setting a lot of RAM aside for the disk cache, I'd expect it to really hum. (Make sure they both have all 32-but memory, too. Just a thought.) > David C. Droman {uunet!attmail}!vsi!droman > V-Systems, Inc. droman@vsi.com > (714) 545-6442 GREEAAT!!! Paul S. R. Chisholm, AT&T Bell Laboratories, att!pegasus!psrc psrc@pegasus.att.com, AT&T Mail !psrchisholm I'm not speaking for the company, I'm just speaking my mind.