*BSD News Article 11257


Return to BSD News archive

Received: by minnie.vk1xwt.ampr.org with NNTP
	id AA1469 ; Tue, 23 Feb 93 14:42:53 EST
Path: sserve!manuel.anu.edu.au!munnari.oz.au!sgiblab!nec-gw!nec-tyo!wnoc-tyo-news!news.u-tokyo.ac.jp!yayoi!tansei1!mhiroshi
From: mhiroshi@tansei.cc.u-tokyo.ac.jp (H. Murakami)
Newsgroups: comp.unix.bsd
Subject: [386bsd] disk cache controller boards.
Message-ID: <4104@tansei1.tansei.cc.u-tokyo.ac.jp>
Date: 15 Feb 93 21:16:48 GMT
Sender: news@tansei.cc.u-tokyo.ac.jp
Organization: Hokkaido Univ. However I am subject to tansei for JUNET.
Lines: 23


IBM PC clones topics:

   I wonder whether the "DISK CACHE CONTROLLER BOARD"
with several MB of RAMs for IDE's or EISA's manages 
the data read/write with 386BSD0.1 correctly.

   The question is, if the disk cache board algorithm 
is not write-through, how the unix periodical sync process 
fullfil its purpose.

Are the board itself does some kind of periodical sync writes actions?
Or, it is on the DOS some kind of cache flushing command is supplied with
or the DOS configured device driver does update of the disk by the 
forcing write at the power off????

I read on magazines, that the disk cache controller does really 
have nice power of improving the performance of the disk I/Os.

Thank you in advance.

	Hiroshi Murakami 
	mhiroshi@tansei.cc.u-tokyo.ac.jp