Re: [Ext2-devel] Re: Shrinking ext3 directories

[Helge Hafting <helgehaf@aitel.hist.no>]

Daniel Phillips wrote:

> I ran a bakeoff between your new half-md4 and dx_hack_hash on Ext2.  As
> predicted, half-md4 does produce very even bucket distributions.  For 200,000
> creates:
> 
>    half-md4:        2872 avg bytes filled per 4k block (70%)
>    dx_hack_hash:    2853 avg bytes filled per 4k block (69%)
> 
> but guess which was faster overall?
> 
>    half-md4:        user 0.43 system 6.88 real 0:07.33 CPU 99%
>    dx_hack_hash:    user 0.43 system 6.40 real 0:06.82 CPU 100%
> 
> This is quite reproducible: dx_hack_hash is always faster by about 6%.  This
> must be due entirely to the difference in hashing cost, since half-md4
> produces measurably better distributions.  Now what do we do?

No surprise that the worse distribution is faster - you get less
io when fewer blocks are used.  Which means a bad distribution beats 
a good one _until_ blocks start to really fill up and collide. 2.8K per
4K block is only 70% full.  I guess the better hash wins
if you force a higher fill rate?

Helge Hafting