Thread: [coLinux-users] Differences of colinux performances between FAT32 and NTFS

SourceForge Headquarters 1320 Columbia Street Suite 310 San Diego, CA 92101 +1 (858) 422-6466

Hi,
somebody knows the differences of performances between installing
colinux on a FAT32 partition and installing colinux on NTFS partition?

thanks

-- 
Giacomo

I've heard a lot of debates either way. I've heard that NTFS has a higher
performance because its been re-written to be more efficient with its
writes, caching, etc. Its also supposedly better because of its better
error-correction (it doesn't need a disk check on boot if you don't properl=
y
shut it down, like FAT32.. due to disk logging like ext3/reiserfs). I've
heard that FAT32 is better because its simpler disk I/O and doesn't have a
lot of the overhead that NTFS does. Plus its more compatible with other OSe=
s
because its not proprietary like NTFS is.

Either way, I've not seen a difference between using coLinux on either.
(Except, of course, for the disk checking on boot with FAT32.) With proper
stats, I can't also say that my experience is definitive, either. I'm sure
it would depend a LOT on hardware.

Ian

On 1/26/06, Giacomo <ky...@ti...> wrote:
>
> Hi,
> somebody knows the differences of performances between installing
> colinux on a FAT32 partition and installing colinux on NTFS partition?
>
> thanks
>
> --
> Giacomo
>
>
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep through log
> files
> for problems?  Stop!  Download the new AJAX search engine that makes
> searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
> http://sel.as-us.falkag.net/sel?cmd=3Dlnk&kid=3D103432&bid=3D230486&dat=
=3D121642
> _______________________________________________
> coLinux-users mailing list
> coL...@li...
> https://lists.sourceforge.net/lists/listinfo/colinux-users
>

Giacomo schrieb:

> somebody knows the differences of performances between installing
> colinux on a FAT32 partition and installing colinux on NTFS partition?

Some benchmark results with bonnie++ follow, according to these it doesn't matter. Maybe the file cache of windows is flattening all differences.

Tests done on a colinux with 200MB, no swap. Windows has 1GB RAM.

ext3 on FAT

Version  1.03       ------Sequential Output------ --Sequential Input- --Random-
                    -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
colinux1      1000M 18065  79 32181  61 14887  68 18573  97 35355  95 189.2  98
                    ------Sequential Create------ --------Random Create--------
                    -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
                 16  1512  99 +++++ +++ +++++ +++  1522  99 +++++ +++  3479  99
colinux1,1000M,18065,79,32181,61,14887,68,18573,97,35355,95,189.2,98,16,1512,99
,+++++,+++,+++++,+++,1522,99,+++++,+++,3479,99

ext3 on NTFS

Version  1.03       ------Sequential Output------ --Sequential Input- --Random-
                    -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
colinux1      1000M 17529  80 32172  60 13995  70 18728  97 35070  96 199.5  97
                    ------Sequential Create------ --------Random Create--------
                    -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
                 16  1519  99 +++++ +++ +++++ +++  1528  99 +++++ +++  3637  99
colinux1,1000M,17529,80,32172,60,13995,70,18728,97,35070,96,199.5,97,16,1519,99
,+++++,+++,+++++,+++,1528,99,+++++,+++,3637,99

ext2 on FAT

Version  1.03       ------Sequential Output------ --Sequential Input- --Random-
                    -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
colinux1      1000M 19003  86 36250  69 15547  74 18765  98 36197  96 217.1  96
                    ------Sequential Create------ --------Random Create--------
                    -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
                 16  2384 100 +++++ +++ +++++ +++  2225  99 +++++ +++  4750 100
colinux1,1000M,19003,86,36250,69,15547,74,18765,98,36197,96,217.1,96,16,2384,100
,+++++,+++,+++++,+++,2225,99,+++++,+++,4750,100

ext2 on NTFS

Version  1.03       ------Sequential Output------ --Sequential Input- --Random-
                    -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
colinux1      1000M 19576  87 36260  72 14557  75 18889  98 35025  96 212.1  97
                    ------Sequential Create------ --------Random Create--------
                    -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
                 16  2325  99 +++++ +++ +++++ +++  2264 100 +++++ +++  4352  99
colinux1,1000M,19576,87,36260,72,14557,75,18889,98,35025,96,212.1,97,16,2325,99
,+++++,+++,+++++,+++,2264,100,+++++,+++,4352,99

Holger Krull ha scritto:
> Some benchmark results with bonnie++ follow, according to these it 
> doesn't matter. Maybe the file cache of windows is flattening all 
> differences.
> 
> Tests done on a colinux with 200MB, no swap. Windows has 1GB RAM.

could you pls repost me the statistics, well aligned with a little
explanation for the data?
For example: %Cp means?

thanks

-- 
Giacomo

Giacomo schrieb:
> Holger Krull ha scritto:
>> Some benchmark results with bonnie++ follow, according to these it 
>> doesn't matter. Maybe the file cache of windows is flattening all 
>> differences.
>>
>> Tests done on a colinux with 200MB, no swap. Windows has 1GB RAM.
> 
> could you pls repost me the statistics, well aligned with a little
> explanation for the data?
> For example: %Cp means?
Amount of processor usage.

Search for bonnie++, every linux has it.
Nice explanation usually in 
/usr/share/doc/bonnie++/readme.html
or
/usr/share/doc/packages/bonnie++/readme.html

Thread: [coLinux-users] Differences of colinux performances between FAT32 and NTFS

Run Linux on Windows or other OSes, natively.

colinux-users