| 1 |
Volker Armin Hemmann <volker.armin.hemmann@××××××××××××.de> wrote: |
| 2 |
|
| 3 |
> > Oh, and according to this benchmark |
| 4 |
> > http://linuxgazette.net/122/piszcz.html |
| 5 |
> > reiserfs does not deserve its speed fame. |
| 6 |
> |
| 7 |
> they tested crap. |
| 8 |
> |
| 9 |
> As I wrote in the other mail. XFS and reiserfs turn on barriers by default, |
| 10 |
> ext3 turns them off. |
| 11 |
> With barriers on for ext3 it looses 30%(!). reiserfs and xfs don't suffer as |
| 12 |
> much, but suffer they do. So if the test did not turn on or off barriers for |
| 13 |
> all fs who support them, ext3 had an unfair advantage. |
| 14 |
> |
| 15 |
> And you want barriers. |
| 16 |
|
| 17 |
I am not sure what you call "barriers".... |
| 18 |
|
| 19 |
ext3 slows down by 400% if you call fsync(2) after copying single files. |
| 20 |
|
| 21 |
UFS on Solaris slows down by 10% only because UFS has been optimized for best |
| 22 |
speed _and_ best data integrity. |
| 23 |
|
| 24 |
Jörg |
| 25 |
|
| 26 |
-- |
| 27 |
EMail:joerg@××××××××××××××××××××××××.de (home) Jörg Schilling D-13353 Berlin |
| 28 |
js@××××××××××××.de (uni) |
| 29 |
schilling@××××××××××××××××.de (work) Blog: http://schily.blogspot.com/ |
| 30 |
URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily |
Archives