Path: utzoo!attcan!uunet!lll-winken!lll-tis!ames!elroy!david
From: david@elroy.Jpl.Nasa.Gov (David Robinson)
Newsgroups: comp.unix.wizards
Subject: Re: NFS block number ordering: how to change?
Summary: biod
Message-ID: <6790@elroy.Jpl.Nasa.Gov>
Date: 19 May 88 22:46:53 GMT
References: <423@cloud9.UUCP>
Organization: Image Analysis Systems Grp, JPL
Lines: 19

In article <423@cloud9.UUCP>, bob@cloud9.UUCP (Bob Toxen) writes:
> We have a Sun running the NFS client side (a kernel implementation).
> When we try to read a file off of the remote system it doesn't request
> blocks 0, 1, 2, 3, ...  Instead it "sorts" them according to the
> disk interlace factor so it actually requests 0, 10, 20, 1, 11, 21, etc.
> I'm sure it's easy to fix this.  I just don't know how?  Could someone
> email me the answer?

What is happening is that the NFS client is attempting to "read-ahead"
the next blocks.  This is done by the "biod"'s.  If you want strictly
increasing from 1 just kill off all of the biod's.  In some cases, this
will actually speed up read speeds with some servers.



-- 
	David Robinson		elroy!david@csvax.caltech.edu     ARPA
				david@elroy.jpl.nasa.gov	  ARPA
				{cit-vax,ames}!elroy!david	  UUCP
Disclaimer: No one listens to me anyway!