[Gluster-users] Strange performance issues on Debian/Lenny compared to Ubuntu

Olivier Le Cam Olivier.LeCam at crdp.ac-versailles.fr
Wed Mar 31 15:54:29 UTC 2010


Paul Kölle wrote :

> Hmm, if I read this right, there is no difference in blocksize:
> 
> (Olivier) Ubuntu (corrected):
> # dd if=/dev/zero of=/mnt/file_test count=262144 bs=1024
> 262144+0 records in
> 262144+0 records out
> 268435456 bytes (268 MB) copied, 5,56942 s, 48,2 MB/s
> 
> (Olivier) Lenny:
> lenny:# dd if=/dev/zero of=/mnt/file_test count=262144 bs=1024
> 262144+0 enregistrements lus
> 262144+0 enregistrements écrits
> 268435456 bytes (268 MB) copied, 82,408 s, 3,3 MB/s
> 
> (mine) Lenny, 3.0.2:
> # dd if=/dev/zero of=/tmp/gluster_mount/file_test count=262144 bs=1024
> 262144+0 records in
> 262144+0 records out
> 268435456 bytes (268 MB) copied, 6.54638 s, 41.0 MB/s
> 
> Im still on kernel 2.6.18 though...

That set me thinking and I can confirm such a difference using either 
2.6.18 or 2.6.26 kernel. Here are the results with both of them:

* Debian/Lenny w/ vernel v2.6.26-2-amd64

     ~# dd if=/dev/zero of=/mnt/test_file count=262144 bs=1024
     262144+0 enregistrements lus
     262144+0 enregistrements écrits
     268435456 bytes (268 MB) copied, 72,2034 s, 3,7 MB/s

* Debian/Lenny w/ kernel v2.6.18-6-amd64 (amd64)

     ~# dd if=/dev/zero of=/mnt/test_file count=262144 bs=1024
     262144+0 enregistrements lus
     262144+0 enregistrements écrits
     268435456 bytes (268 MB) copied, 7,53942 s, 35,6 MB/s

As you can see, dd'ing is about 10 times faster with kernel 2.6.18!

Anyway, postmark benching is quite equivalent regardless the kernel 
version so it does not really make me satisfied for a webservers farm! :)

PS: the client machine I used here for these tests is an old Pentium(R) 
D 2.80GHz with 2GBytes of RAM.

Just in case it can help, thought.

Thanks and best regards,
-- 
Olivier



More information about the Gluster-users mailing list