read performance much slower than write

Hi!

I am running enbd with 4 server builing raid 5 on the client with 4 devices.
I build raid using mdadm and everything works fine so far. Raid device is
formated with reiserfs.

Now I started testing the performance using bonnie++ and discovered that the
write performance is much higher than the read performance!
Here is the output of bonnie++:

bonnie++ -d /mnt/raid5/ -s 4096 -n 0 -f -u root

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
enbd-client26  4G           	    72113  42    18330   9
11445   3      755.9   2
enbd-client26	4G,,,72113,42,18330,9,,,11445,3,755.9,2,,,,,,,,,,,,,

As you can see the write performance is very high, the read performance is
even worse than the rewrite.

I am starting the enbd-server using enbd.conf:
server   backend1  1099 /dev/sdc1 -b 4096
server   backend2  1199 /dev/sdc1 -b 4096
server   backend3  1299 /dev/sdc1 -b 4096
server   backend4  1399 /dev/sdc1 -b 4096

The client also using enbd.conf:
client   backend1  /dev/nda 192.168.100.185 1099 -n 2            -e -m -b
4096
client   backend2  /dev/ndb 192.168.100.186 1199 -n 2            -e -m -b
4096
client   backend3  /dev/ndc 192.168.100.187 1299 -n 2            -e -m -b
4096
client   backend4  /dev/ndd 192.168.100.188 1399 -n 2            -e -m -b
4096

I already tried to echo "merge_requests=512" > /proc/nbdinfo which didn´t
change much.

Shouldn´t be the performance of read operations better or at least almost
the same than the write performance?
Am I missing spezial parameter on the client or server side?

Thank´s for helping me out,

Christian