Hi,
I have two server nodes, gluster1 & gluster2 ,running on Ubuntu 14.04, and is using gluster ver 3.7.6 . I am using fio tools ver 2.6 for testing performance.(IO engine= libaio)
gluster1 server node is a high performance CPU (8 core , i7/3.4GHz) with SSD (random rw performance without glusterFS is ~14MB/s with ext4 FS, 150 GB ), while
gluster2 is a Pentium g840 CPU (2 core, 2.8GHz) with a WD HDD of 1 TB size.(random rw performance without glusterFS is ~500KB/s only with ext4).
1. I see gluster1 seq read performance is much higher than gluster 2. See below data(in KB/s).Is this because gluster1 CPU is better and it is using SSD ?(replica 2)
| Seq. read | Seq. write | |
| gluster1 | 319882 | 83253 |
| gluster2 | 135931 | 61059 |
2. I also see gluster2 performs well when configured as distributed volumes, compared to gluster1. Why is it so ? Pls see the below data.(in KB/s)
| random read | random write | ||
| gluster1 | 1029 | 1027 | |
| gluster2 | 3358 | 3353 | |
Expecting experts' answer.
Best Regards,
JK
_______________________________________________ Gluster-users mailing list Gluster-users@xxxxxxxxxxx http://www.gluster.org/mailman/listinfo/gluster-users
