lkml.org 
[lkml]   [1998]   [Jun]   [16]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: e2fs + large disk == slug?
On Tue, 16 Jun 1998, Mark Lehrer wrote:
> On Mon, 15 Jun 1998, Mark Lehrer wrote:
>
> > 1) removing 500 megs of files took 30 seconds.
>
> This will be just as slow on a 600M filesystem...
> If there are a lot of files, it's going to take a rather
> long time to remove them. (Just try this under Windows:)
>
> On my 120mhz HP-UX system this same operation on the same set of files
> with the same model disk drive only takes a few seconds; whereas with
> my 200MHz PPro Linux system it takes five to ten times as long...

This is because Linux does synchronous metadata updates on
unlink(). And even when it doesn't do that, it will still
need to do synchronous metadata _reads_ in order to determine
what to write out asynchronously.

> Is it possible to optimize this?

Well, we could recode the kernel to let bdflush or a new
kernel daemon do the unlinks... Apart from that, there's
not much of a solution :(
(or you simply run 'rm -rf dir &', which will probably
be the same thing that Solaris does internaly :-)

Rik.
+-------------------------------------------------------------------+
| Linux memory management tour guide. H.H.vanRiel@phys.uu.nl |
| Scouting Vries cubscout leader. http://www.phys.uu.nl/~riel/ |
+-------------------------------------------------------------------+

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu

\
 
 \ /
  Last update: 2005-03-22 13:43    [from the cache]
©2003-2011 Jasper Spaans