[Sluglug] new to the list

[Peter Belew]

Hmm ...

That's basically what I would think - the main advantage of
having the files spread out is to reduce seek time to an
empty area on the disk.

That assumes that the disk structure info - FATs, INODE
tables, whatever the file system uses, needed to find
what's available  - is cached in memory so you don't have
 to seek to it. (But then you have to write it back soon
anyway).

The XEROX PARC ALTO file system avoided some of that by
having a separate header on each block with previous/next
pointers (and a file-ID/free field) - you didn't need to access
any other data structure to read files (once opened) and
some writing could be done without looking at a free block
bitmap - though that had to be updated at some point.

Peter

On 10/3/05, Ignacio Solis <isolis at igso.net> wrote:
> * cerise at armory.com (cerise at armory.com) said:
> > Either way, the frequency of use would be interesting, but you wouldn't
> > want to put all those files together.  You'd try to spread those out over the
> > platter so that at any one time, you're near a hot file.
>
> That doesn't make sense. If you want fast access (seek time) you want all the
> files close to you (the head), hence you move them close together.  You'll
> achieve nothing if you spread them out... other than increasing the seek time
> from one file to another.
>
> Nacho
>
> --
> "In Googlis non est, ergo non est." - Anonymous Coward
> Homepage:       http://www.cse.ucsc.edu/~isolis/  |  EEE8 08C9 FBAE B471 9691
> GPG Public Key: http://www.igso.net/isolis.gpg    |  CE7A 1CC8 D3DE B31E 10AB
>
>
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.1 (GNU/Linux)
>
> iD8DBQFDQaSlHMjT3rMeEKsRAvmrAJ0UG1E8a2elon3lB53w8rWM+xSQSQCeOiYG
> BYK+P7Y+IyuUJgvrMe5hZXA=
> =GjLO
> -----END PGP SIGNATURE-----
>
>
> _______________________________________________
> Sluglug mailing list
> Sluglug at sluglug.ucsc.edu
> http://sluglug.ucsc.edu/mailman/listinfo/sluglug
>
>
>


--
http://www.sonic.net/~peterbe/
Email: peterbe at sonic.net