FS statistics on 2.6.0-test2
Net Llama!
netllama
Mon May 17 11:50:46 PDT 2004
On 08/08/03 06:20, Collins Richey wrote:
> Interesting article and lots of critiques at
>
> http://kerneltrap.org/node/view/715
>
> The following extract shows the basic differences. This is German
> number notation with commas and periods reversed.
>
> reiser4 reiserfs ext3 XFS JFS
> copy 33.39,34% 39.55,32% 39.42,25% 43.50,32% 48.15,20%
> sync 1.54,00% 3.15,10% 9.05,00% 2.08,10% 3.05,10%
> recopy1 31.09,34% 75.15,13% 79.96, 9% 102.37,12% 108.39, 5%
> recopy2 33.15,33% 77.62,13% 98.84, 7% 108.00,12% 114.96, 5%
> sync 2.89, 3% 3.84, 1% 8.15, 0% 2.40, 2% 3.86, 0%
> du 2.05,42% 2.46,21% 3.31,11% 3.73,32% 2.42,17%
> delete 7.41,52% 5.22,58% 3.71,39% 8.75,56% 15.33, 7%
> tar 52.25,25% 90.83,12% 74.93,13% 157.61, 7% 135.86, 6%
> sync 6.77, 2% 4.19, 3% 1.67, 1% 0.95, 1% 38.18, 0%
> overall 171.28,30% 302.53,16% 319.71,11% 429.79,13% 470.88, 6%
>
> In general, reiser4 looks to be a real winner (when it's out of
> development, of course), and my favorite EXT3 and everyone else's
> favorite XFS are among the slowest.
Its also worth noting that there have been quite a few discussions about
this on the XFS list. The developers there have already poked a few holes
in the accuracy of these measurements, such as the fact that comparing
anything on a 2.6.0-test kernel is pointless considering how much of a
moving target it is, plus the distinct lack of information on the version
of xfs userland tools in use. Also, XFS in 2.6.x is not considered to be
stable by the XFS developers, and in their words, "anyone using 2.6.0-test
kernels for anything serious needs to have their head examined".
What would be far more telling would be a similar test on 2.4.21, with
detailed information on the test environment.
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
L. Friedman netllama at linux-sxs.org
Linux Step-by-step & TyGeMo: http://netllama.ipfox.com
6:55am up 24 days, 9:37, 2 users, load average: 0.40, 0.23, 0.10
More information about the Linux-users
mailing list