Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!sol.ctr.columbia.edu!lll-winken!elroy.jpl.nasa.gov!decwrl!ucbvax!NISC.SRI.COM!cwilson
From: cwilson@NISC.SRI.COM (Chan Wilson)
Newsgroups: comp.sys.apple
Subject: Re: file system questions
Message-ID: <13373@fs2.NISC.SRI.COM>
Date: 19 Feb 90 23:32:11 GMT
References: <1557@crash.cts.com> <38736@apple.Apple.COM> <3637@sage.cc.purdue.edu>
Reply-To: cwilson@NISC.SRI.COM (Chan Wilson)
Organization: Network Info Systems Ctr., SRI Intl., Menlo Park, CA.
Lines: 39

>>In article <1557@crash.cts.com> tsouth@pro-pac.cts.com writes:
>>>
[much ado over 32meg limitation of Prodos8]
>    There's a Sierra-Online FST.  This allows those who have expensive
	       ^^^^^^^^^^^^^
Hmm, that's an amusing typo. ;-)

[Description of various FSTs]

>    So Matt, you say ProDOS 16 and GS/OS were written to get around the
>limitations of ProDOS 8?  The only storage capicaty limit that I see as
>being removed is the number of volumes per slot.  With the size of hard
>drives getting larger, we still CANNOT make good use of them without
>having to resort to defining many partions to fill the space.  This is
>another main reason why we want a mac FST, so we can format the hard
>drive to one big volume!  Some will even settle for a FST of a new file
>system, one that will permit GS/OS to live up to its fullest protentials!

Ever realize that HFS is limited to 32 MB volumes?  Betcha you didn't
know that. :) HFS gets around the limitation by using allocation
blocks in conjunction with logical blocks to achieve the larger volume
size.  Catch is, there's a tradeoff.  The larger the drive is, the
larger the allocation blocks are.  This means that every time you
cross a multiple of 32 in size, your a-block size goes up by another
512 bytes.  The extreme end of this sees you with a volume 256
Terrabytes long, with a minimum file size of 4 gigabytes.  I wonder
what the access time of a drive that big would be?  ;-)

Now, I'm not that familiar with GS/OS (yet), but it seems that some
possible variation in this theme would be possible.  Question is, when?

> Y_,_|[]|   Eric Mulholland

--Chan
			   ................
  Chan Wilson -- cwilson@nisc.sri.com <or> radius!cwilson@apple.com
Janitor/Architect of comp.binaries.apple2 archive on wuarchive.wustl.edu
	      I don't speak for SRI, someone else does.
			   ................