[Gluster-users] Performance problems

[Aurélien ROUGEMONT]

Le 29/03/10 18:48, Olivier Le Cam a écrit :
> Bonjour Aurélien,
>
> Aurelien ROUGEMONT wrote:
>> root at test4:/var/www# dd if=/dev/zero of=./test count=10000 bs=1024
>> 10000+0 records in
>> 10000+0 records out
>> 10240000 bytes (10 MB) copied, 17.8632 s, 573 kB/s
>
> I have posted the exact same results just a few days ago (using the 
> sid package backported to lenny):
> http://gluster.org/pipermail/gluster-users/2010-March/004240.html
>
> BTW, this seems to be related to the block size. Performances might be 
> very different with different block sizes:
>
> gl2:/web-nunki# dd if=/dev/zero of=./test count=10000 bs=2000
> 10000+0 enregistrements lus
> 10000+0 enregistrements écrits
> 20000000 bytes (20 MB) copied, 3,25723 s, 6,1 MB/s
>
> gl2:/web-nunki# dd if=/dev/zero of=./test count=10000 bs=10000
> 10000+0 enregistrements lus
> 10000+0 enregistrements écrits
> 100000000 bytes (100 MB) copied, 3,30255 s, 30,3 MB/s
>
> gl2:/web-nunki# dd if=/dev/zero of=./test count=10000 bs=40000
> 10000+0 enregistrements lus
> 10000+0 enregistrements écrits
> 400000000 bytes (400 MB) copied, 5,97635 s, 66,9 MB/s
>
> gl2:/web-nunki# dd if=/dev/zero of=./test count=10000 bs=80000
> 10000+0 enregistrements lus
> 10000+0 enregistrements écrits
> 800000000 bytes (800 MB) copied, 9,12432 s, 87,7 MB/s
>
> gl2:/web-nunki# dd if=/dev/zero of=./test count=10000 bs=160000
> 10000+0 enregistrements lus
> 10000+0 enregistrements écrits
> 1600000000 bytes (1,6 GB) copied, 15,0147 s, 107 MB/s
>
> gl2:/web-nunki# dd if=/dev/zero of=./test count=10000 bs=320000
> 10000+0 enregistrements lus
> 10000+0 enregistrements écrits
> 3200000000 bytes (3,2 GB) copied, 28,8645 s, 111 MB/s
>
> TBH, I don't know what to conclude from these results.
>
> Regards,
Bonsoir Olivier,

I did not have the time to write a proper addendum to my earlier emails. 
I came to the very same conclusions. And I am  now in front of some 
questions like :

will this be ok for a web servers farm ?
what dies it mean ? What is the most common bs= equivalent for a www 
server ? Do i have to change the block size of the underlying filesystem ?

(by the way, i've managed to get the same results with an opensolaris 
2009.06 + glusterfs 3.0.3)

I would be glad to get some more explanations about this

Thank you all for taking the time to answer to my wonderings.

Best regards,

Aurélien


!DSPAM:4bb12b3b212821458925448!