lkml.org 
[lkml]   [2004]   [Apr]   [14]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [Ext2-devel] Re: [RFC] extents,delayed allocation,mballoc for ext3
From
Date
On Wed, 2004-04-14 at 05:10, Alex Tomas wrote:
>
> I've just benched ext3 vs. ext3+reservation vs. ext3+delalloc vs. xfs.
> it was tiobench.
> ext3 1024 4096 32 8.12 9.872% 8.111 82
> ext3-dalloc 1024 4096 32 24.83 20.01% 2.995 124
> ext3-reserv 1024 4096 32 22.72 29.51% 3.282 77
> xfs 1024 4096 32 25.47 21.75% 2.247 117
>
Hi Alex,

Nice comparison! The ext3 reservation system use more cpus because we do
reservations in memory( not on disk) and we have a global lock per
filesystem to guard the operation. The current search for a new
reservation window algorithm is not perfect right now.

extents and delayed allocation probably is the right way to go for next
generation (maybe ext4). Currently I just try to fix the missing
preallocation feature in ext3, without break the disk compatibility and
involve too much changes....

Thanks for your interest.

Mingming

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 14:02    [W:0.088 / U:1.056 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site