[Gluster-users] read performance

[Alex Leung]

Can someone explain why there are a big read performance difference {ls -l)
on a glusterfs mount-point than on the bricks mount-point?

[root at pdsraid5 ~]# gluster volume info

Volume Name: gv0
Type: Replicate
Volume ID: b3052157-b234-45a6-b4f4-ceacf9390567
Status: Started
Number of Bricks: 1 x 2 = 2
Transport-type: tcp
Bricks:
Brick1: pdsraid5:/export/pdsdata1
Brick2: pdsmirror1:/export/pdsdata1
[root at pdsraid5 ~]#


[root at pdsraid5 ~]# df -h
Filesystem            Size  Used Avail Use% Mounted on

/dev/sdb1              19T  1.3T   18T   7% /export/pdsdata1
pdsraid5:/gv0          19T  1.3T   18T   7% /export/pdsdata

[root at pdsraid5 ]# time ls -lR /export/pdsdata1/marci

real	0m0.465s
user	0m0.126s
sys	0m0.337s


[root at pdsraid5 ]# time ls -lR /export/pdsdata/marci
real	0m37.429s
user	0m0.228s
sys	0m1.031s
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://supercolony.gluster.org/pipermail/gluster-users/attachments/20130918/e81892a7/attachment.html>