Return to BSD News archive
Path: sserve!manuel!munnari.oz.au!uunet!ukma!darwin.sura.net!zaphod.mps.ohio-state.edu!moe.ksu.ksu.edu!maverick.ksu.ksu.edu!news From: brtmac@maverick.ksu.ksu.edu (Brett McCoy) Newsgroups: comp.unix.bsd Subject: Re: 386BSD and IDE drives Date: 23 Sep 1992 19:52:57 -0500 Organization: Kansas State University Lines: 28 Message-ID: <19r3h9INN6fe@maverick.ksu.ksu.edu> References: <1992Sep22.224636.13727@spang.Camosun.BC.CA> <1992Sep23.210231.15155@fcom.cc.utah.edu> NNTP-Posting-Host: maverick.ksu.ksu.edu terry@cs.weber.edu (A Wizard of Earth C) writes: >Some people argue that a lot of what UNIX does in terms of caching isn't >really needed any more with caching controllers. I don't think this is >going to be true until the data transfer rate from the controller over the >bus is the same as that to "local" (motherboard) memory. Until then, soft >caching should continue to be higher performance. >The bad news is that there are starting to be SCSI drives with translation, >and pretty soon there may be no safe disk technology. My understanding is that geometry tranlation only hurts writes, not reads, on most IDE drives since most, like MAXTOR, have cache's on the onboard controller so that the drive can always read the entire track being read and then fulfill susequent requests to the same track from that cache. This doesn't help writes since these caches are write thru, not write back. Also, many of the LARGE SCSI drives are now using gemetry translation because not all of the cylinders have the same number of sectors. In order to get the most out of the media there are more sectors on the outer cylinders and fewer on the inner cylinders. There is no way to do this effectively other than geometry mapping. Now, drives with battery backed write-back caches would be great. Then the machine accessing the disk doesn't need to do anything fancy to get the most out of the drive. ++Brett;