Re: Re: [STUPID TESTCASE] ext3 htree vs. reiserfs on 2.5.40-mm1

[Hans Reiser <reiser@namesys.com>]

Dieter Nützel wrote:

>Am Mittwoch, 2. Oktober 2002 10:48 schrieb Paul P Komkoff Jr:
>  
>
>>Replying to Andreas Dilger:
>>    
>>
>>>Why do you think data=writeback is better than data=journal?  If the
>>>files have no data then it should not make a difference.
>>>      
>>>
>>It is better than default data=ordered I think :)
>>
>>Thanks for detailed explanation - it saved much time for me and
>>accortind to yours directions I have recalculated my test. Now ext3 is
>>better :)
>>
>>e3
>>create          2m49.545s       0m4.162s        2m20.766s
>>delete          2m8.155s        0m3.614s        1m34.945s
>>
>>reiser
>>create          3m13.577s       0m4.338s        2m54.026s
>>delete          4m39.249s       0m3.968s        4m16.297s
>>
>>e3
>>create          2m50.766s       0m4.024s        2m21.197s
>>delete          2m8.755s        0m3.501s        1m35.737s
>>
>>reiser
>>create          3m13.015s       0m4.432s        2m53.412s
>>delete          4m41.011s       0m3.893s        4m16.845s
>>
>>
>>this is two typical runs. Now I creating ext3 with
>>mke2fs -j -O dir_index -J size=192 -T news /dev/sda4
>>
>>as you can see, this improves performance by 1/4
>>
>>Unfortunately, there still one issue in ext3. It called "inode limit".
>>Initially I wanted to run this test on 1000000 files but ... I hit
>>inode limit and don't want to increase it artificially yet.
>>
>>Reiserfs worked fine because it don't have such kind of limit ...
>>    
>>
>
>Orange and apples, again?
>
>Try with latest ReiserFS data-logging and see what we get?
>Chris do you have something for 2.5.40+ ready?
>
>2.5.40 is GREAT!
>
>Regards,
>	Dieter
>  
>
So, I think that if you create enough files in your test to stress the 
IO subsystem, then it will be a much more interesting test. 
 Unfortunately, you may need to increase the inode allocation to test 
this for ext3.  Once you start to get above a million files, you may 
start to see reiser4 doing substantially better than ReiserFS because 
reiser4 has better layout and better hashes based on things we learned.

I would be curious to see your benchmark done for 3 million files.  I 
would be curious to see what happens if you umount in the middle of it. 
 I would not be surprised if interesting VFS/dcache performance issues 
arise as you increase the number of files.

Hans