On Thu, 2006-03-30 at 12:32, UNIX admin wrote:

> I would strongly discourage switching from bzip2. bzip2 may be "slow", but 
> CPUs will only get faster, and meanwhile, the compression algorithm of bzip2
> will stay the same.
> 
> When one factors in that no other cruncher/compressor/packer/archiver gets
> that kind of compressions, I believe that long term gains become pretty 
> obvious.

Not really. The fastest systems today can't saturate a DVD, 100M
network, or hard
disk with bzip2.

I would happily exchange the 5% loss in compression for the 10x
performance win.

If you need that 5%, then there are probably other ways to get it back.
The ISO
images compress (with zip) 20% or so - and that's with the packages
bzip2 compressed.

> > 3) stop rewriting /var/sadm/install/contents file
> > repeatedly.
> 
> Shouldn't that already be eleviated by FS cache in RAM and logging?

The cache only works for reads; logging only for metadata.

(Casper said it's "transactionally safe". I would think that the
relevant
transactional unit here is the whole install.)

> > If we stored the packages as a single CPIO archive
> > instead
> > of separate files on the dvd, we could stream the
> > data off
> > continuously, uncompress it and install it all at the
> > same
> > time.
> 
> That might work.  But what happens when that pesky user selects a custom
> installation and thereby screws up the `cpio` package dearchiving order?

That's not how I understood it. I thought the idea was that the package
would be stored as a single file in datastream format rather than
unpacked
in filesystem format. Which would cut down a lot of the small I/O
operations
that are currently necessary.

-- 
-Peter Tribble
L.I.S., University of Hertfordshire - http://www.herts.ac.uk/
http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/


_______________________________________________
opensolaris-discuss mailing list
[email protected]

Reply via email to