There are many very good reasons its the new default filesystem in RHEL 7
Also SSDs are faster at random IO and small file; however a properly built RAID of spinning disks is still faster at linear reads of large file.
In general qemu does large linear reads or at least very close to it of the VM's virtual disks on boot.
-- Sent from my HP Pre3
On Nov 3, 2014 6:17 PM, Lindsay Mathieson <lindsay.mathieson@xxxxxxxxx> wrote:
I hope I'm not spamming and irritating the list with all this, my
apologies if its all old hat.
I recreated my gluster 2 disk replica store with ext4 using a 5GB SSD
journal for each journal. So far it seems to be a huge improvement.
VM start up's still trigger 5-10% iowait, but I presume that sustained
reads which the cache wouldn't help so much with.
However general operation of 5 VDI windows vm's (Developer machines)
is much improved with wait times averaging round 0-3% and over
responsiveness very good.
Which leads me to my inevitable questions :)
- Journal size. I actually have up to 55GB of SSD spare. Is there any
point in increasing the SSD journal size over 5GB? the ext4 disk is
3TB (Western Digital Red)
- ext4 creation options. Any recommenced options there?
- ext4 mount options. Recommend options? I'm using "noatime,nodiratime"
- Would it be a real bad idea to mount with:
* write back cache enabled
* nobarrier
At the moment this is a test system so I can rebuild without issues,
though copying the test VM's takes a few hours.
thanks,
--
Lindsay
_______________________________________________
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
http://supercolony.gluster.org/mailman/listinfo/gluster-users
_______________________________________________ Gluster-users mailing list Gluster-users@xxxxxxxxxxx http://supercolony.gluster.org/mailman/listinfo/gluster-users
