On Tue, 2 Dec 1997 17:00:23 -0800  Matthew Jacob wrote:

> 
> >8k per inode?  I've got a 2GB filesystem here with over 100k inodes and
> >1.5G used.  This proposal would chew up around 800MB, or about 40% of
> >the available space.  Disk space _isn't_ that cheap.
> 
> Uh, at NASA/Ames it is. Currently 4TB is the disk cache. That will more
> than double.

Hmm.  I'll treat Ames as an exception :)

> 
> >
> >Maybe if a finite amount of data was kept in the inode too this wouldn't
> >be as bad.  Something "simpler" might be to just allocate another
> >fragment and use that for "funny" data _if needed_.
> 
> That's another idea. Again, though, this becomes a second access.

I guess this depends on would be stored in the extra space.  Under some
(or a lot of, or most?) circumstances I guess the second access wouldn't
be needed.

Yet another idea (if 128 bytes of extra info is enough (ie the size of
an inode)) would be to change the ufs layer to start allocating inodes
at every second inode initially, so that the second one could "attached"
to the first and that wouldn't require another access.  Then you'd only
get second access problems as the filesystem filled up.  This could be
extended to every forth, eighth, ...

> >ps: should this have stayed on developers and not migrated to tech-kern?
> >Or the other way around?
> >
> 
> We were told to move to tech-kern.

Sorry, missed that one...

Simon.